{
  "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": 0.04467606,
      "auxiliary_loss_mlp": 0.02214109,
      "balance_loss_clip": 2.44035244,
      "balance_loss_mlp": 1.7477622,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 39.94680766841656,
      "language_loss": 2.58248568,
      "learning_rate": 0.0,
      "loss": 1.9010365,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 15.553298234939575
    },
    {
      "auxiliary_loss_clip": 0.02957249,
      "auxiliary_loss_mlp": 0.01419526,
      "balance_loss_clip": 1.61343002,
      "balance_loss_mlp": 1.13208902,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 54.79875405706253,
      "language_loss": 1.88785017,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.93161786,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.7612693309783936
    },
    {
      "auxiliary_loss_clip": 0.02960445,
      "auxiliary_loss_mlp": 0.01484531,
      "balance_loss_clip": 1.62834692,
      "balance_loss_mlp": 1.18774796,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 40.83861379082411,
      "language_loss": 1.61476827,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.65921807,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.541330575942993
    },
    {
      "auxiliary_loss_clip": 0.02935145,
      "auxiliary_loss_mlp": 0.01464239,
      "balance_loss_clip": 1.61329925,
      "balance_loss_mlp": 1.15295982,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 37.01716318323993,
      "language_loss": 1.64450955,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.6885035,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.611107349395752
    },
    {
      "auxiliary_loss_clip": 0.02953273,
      "auxiliary_loss_mlp": 0.01455373,
      "balance_loss_clip": 1.6154902,
      "balance_loss_mlp": 1.15248573,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 55.07727517249346,
      "language_loss": 1.93506944,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.97915578,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.8608996868133545
    },
    {
      "auxiliary_loss_clip": 0.02967778,
      "auxiliary_loss_mlp": 0.01440656,
      "balance_loss_clip": 1.61950994,
      "balance_loss_mlp": 1.14978492,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 45.570967847641434,
      "language_loss": 1.6095438,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.65362811,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.8553831577301025
    },
    {
      "auxiliary_loss_clip": 0.029676,
      "auxiliary_loss_mlp": 0.01921181,
      "balance_loss_clip": 1.74828148,
      "balance_loss_mlp": 1.68619573,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.605888270912766,
      "language_loss": 0.81467319,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.86356103,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.397385358810425
    },
    {
      "auxiliary_loss_clip": 0.0296086,
      "auxiliary_loss_mlp": 0.01508198,
      "balance_loss_clip": 1.61705494,
      "balance_loss_mlp": 1.19100642,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 41.119438373742106,
      "language_loss": 1.58188546,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.62657607,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.84830641746521
    },
    {
      "auxiliary_loss_clip": 0.02948974,
      "auxiliary_loss_mlp": 0.01465889,
      "balance_loss_clip": 1.61906743,
      "balance_loss_mlp": 1.15956855,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 18.08311959645327,
      "language_loss": 1.47888017,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52302885,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.86454701423645
    },
    {
      "auxiliary_loss_clip": 0.02937955,
      "auxiliary_loss_mlp": 0.01467086,
      "balance_loss_clip": 1.62112153,
      "balance_loss_mlp": 1.1544714,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.179708289285298,
      "language_loss": 1.53100634,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57505679,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.8662192821502686
    },
    {
      "auxiliary_loss_clip": 0.02940062,
      "auxiliary_loss_mlp": 0.01492166,
      "balance_loss_clip": 1.61530626,
      "balance_loss_mlp": 1.18927872,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 15.833724520621283,
      "language_loss": 1.32358086,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.36790323,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.8281431198120117
    },
    {
      "auxiliary_loss_clip": 0.02961133,
      "auxiliary_loss_mlp": 0.01485069,
      "balance_loss_clip": 1.61891055,
      "balance_loss_mlp": 1.17684102,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 12.324471154923419,
      "language_loss": 1.25385261,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.29831457,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.96876859664917
    },
    {
      "auxiliary_loss_clip": 0.02941593,
      "auxiliary_loss_mlp": 0.01451554,
      "balance_loss_clip": 1.62169218,
      "balance_loss_mlp": 1.15419841,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.6576706024290715,
      "language_loss": 1.2902782,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.33420968,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.9746713638305664
    },
    {
      "auxiliary_loss_clip": 0.02915523,
      "auxiliary_loss_mlp": 0.01463951,
      "balance_loss_clip": 1.61114192,
      "balance_loss_mlp": 1.16259003,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 8.075965978215015,
      "language_loss": 1.13560903,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.17940378,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 4.783610582351685
    },
    {
      "auxiliary_loss_clip": 0.02905543,
      "auxiliary_loss_mlp": 0.01498061,
      "balance_loss_clip": 1.61611676,
      "balance_loss_mlp": 1.19345713,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 5.534810172583667,
      "language_loss": 1.12852788,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17256391,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 3.828604221343994
    },
    {
      "auxiliary_loss_clip": 0.02850675,
      "auxiliary_loss_mlp": 0.0173106,
      "balance_loss_clip": 1.72495747,
      "balance_loss_mlp": 1.50980794,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.824813040755893,
      "language_loss": 0.64701551,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.69283289,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 3.269314765930176
    },
    {
      "auxiliary_loss_clip": 0.02873206,
      "auxiliary_loss_mlp": 0.01431816,
      "balance_loss_clip": 1.61367083,
      "balance_loss_mlp": 1.13770235,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 4.4468085220211835,
      "language_loss": 1.16141224,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20446253,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 3.0195155143737793
    },
    {
      "auxiliary_loss_clip": 0.02807219,
      "auxiliary_loss_mlp": 0.01668484,
      "balance_loss_clip": 1.71573555,
      "balance_loss_mlp": 1.45180929,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.5637551802310266,
      "language_loss": 0.72239363,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.76715064,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.2330667972564697
    },
    {
      "auxiliary_loss_clip": 0.028581,
      "auxiliary_loss_mlp": 0.01473629,
      "balance_loss_clip": 1.60797453,
      "balance_loss_mlp": 1.16616392,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 5.052849558413146,
      "language_loss": 1.01752734,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.06084466,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.939197063446045
    },
    {
      "auxiliary_loss_clip": 0.02761703,
      "auxiliary_loss_mlp": 0.01591906,
      "balance_loss_clip": 1.70997667,
      "balance_loss_mlp": 1.38133514,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.599382686018435,
      "language_loss": 0.76506174,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.8085978,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 3.0653860569000244
    },
    {
      "auxiliary_loss_clip": 0.02830854,
      "auxiliary_loss_mlp": 0.01447051,
      "balance_loss_clip": 1.60585761,
      "balance_loss_mlp": 1.14569008,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 5.087308143485803,
      "language_loss": 1.01554823,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.05832732,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.815840244293213
    },
    {
      "auxiliary_loss_clip": 0.02784108,
      "auxiliary_loss_mlp": 0.01413503,
      "balance_loss_clip": 1.59741044,
      "balance_loss_mlp": 1.12454009,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 4.875791688057386,
      "language_loss": 0.89484751,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.93682373,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 2.943260431289673
    },
    {
      "auxiliary_loss_clip": 0.02752024,
      "auxiliary_loss_mlp": 0.01386608,
      "balance_loss_clip": 1.58558714,
      "balance_loss_mlp": 1.09802604,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.456878153891676,
      "language_loss": 1.02051497,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.06190133,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.8832197189331055
    },
    {
      "auxiliary_loss_clip": 0.0273319,
      "auxiliary_loss_mlp": 0.01436626,
      "balance_loss_clip": 1.58454442,
      "balance_loss_mlp": 1.14499235,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.14335656242248,
      "language_loss": 1.02702713,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.06872535,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.855179786682129
    },
    {
      "auxiliary_loss_clip": 0.0269432,
      "auxiliary_loss_mlp": 0.01383565,
      "balance_loss_clip": 1.57428026,
      "balance_loss_mlp": 1.10146832,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 4.03219827219884,
      "language_loss": 0.92548281,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.96626163,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 2.9545159339904785
    },
    {
      "auxiliary_loss_clip": 0.02691766,
      "auxiliary_loss_mlp": 0.01389614,
      "balance_loss_clip": 1.5795958,
      "balance_loss_mlp": 1.10656321,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.409555234360913,
      "language_loss": 0.92906344,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.96987718,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.8401596546173096
    },
    {
      "auxiliary_loss_clip": 0.02665196,
      "auxiliary_loss_mlp": 0.01402198,
      "balance_loss_clip": 1.57476282,
      "balance_loss_mlp": 1.11285329,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.2260456861779807,
      "language_loss": 1.00941598,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05008996,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.82129168510437
    },
    {
      "auxiliary_loss_clip": 0.0261853,
      "auxiliary_loss_mlp": 0.01387884,
      "balance_loss_clip": 1.5706892,
      "balance_loss_mlp": 1.10006499,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 3.964031273736502,
      "language_loss": 0.90706223,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.94712639,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.7319085597991943
    },
    {
      "auxiliary_loss_clip": 0.0261848,
      "auxiliary_loss_mlp": 0.0137853,
      "balance_loss_clip": 1.56292892,
      "balance_loss_mlp": 1.09528923,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.344530787211221,
      "language_loss": 0.99772131,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.03769159,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.8428821563720703
    },
    {
      "auxiliary_loss_clip": 0.02608166,
      "auxiliary_loss_mlp": 0.01400786,
      "balance_loss_clip": 1.56078279,
      "balance_loss_mlp": 1.11888003,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.0627046631054524,
      "language_loss": 0.99542361,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.03551304,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.824864149093628
    },
    {
      "auxiliary_loss_clip": 0.02560761,
      "auxiliary_loss_mlp": 0.01364637,
      "balance_loss_clip": 1.55760372,
      "balance_loss_mlp": 1.10943365,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 2.4527429912715255,
      "language_loss": 1.00216627,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04142022,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.8619816303253174
    },
    {
      "auxiliary_loss_clip": 0.02507337,
      "auxiliary_loss_mlp": 0.01361767,
      "balance_loss_clip": 1.54143584,
      "balance_loss_mlp": 1.09607363,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.644670731490008,
      "language_loss": 0.99069548,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.02938652,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 3.1506917476654053
    },
    {
      "auxiliary_loss_clip": 0.02454353,
      "auxiliary_loss_mlp": 0.0135281,
      "balance_loss_clip": 1.52698088,
      "balance_loss_mlp": 1.08501875,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 2.2931215889912235,
      "language_loss": 1.02416885,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06224036,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 3.0651557445526123
    },
    {
      "auxiliary_loss_clip": 0.02413018,
      "auxiliary_loss_mlp": 0.01346079,
      "balance_loss_clip": 1.51708829,
      "balance_loss_mlp": 1.07981277,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 2.7479262377500273,
      "language_loss": 0.881145,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.91873598,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 3.0397989749908447
    },
    {
      "auxiliary_loss_clip": 0.02330016,
      "auxiliary_loss_mlp": 0.01319568,
      "balance_loss_clip": 1.49565673,
      "balance_loss_mlp": 1.06455588,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.1112514141812997,
      "language_loss": 1.03360033,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.07009625,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.849703311920166
    },
    {
      "auxiliary_loss_clip": 0.02318,
      "auxiliary_loss_mlp": 0.01340056,
      "balance_loss_clip": 1.49303532,
      "balance_loss_mlp": 1.07779574,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 2.5247454512658307,
      "language_loss": 0.9506346,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.98721528,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.8934669494628906
    },
    {
      "auxiliary_loss_clip": 0.01970263,
      "auxiliary_loss_mlp": 0.01294235,
      "balance_loss_clip": 1.45352697,
      "balance_loss_mlp": 1.1355437,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.3827576621663105,
      "language_loss": 0.60941261,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.6420576,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.2487406730651855
    },
    {
      "auxiliary_loss_clip": 0.02230809,
      "auxiliary_loss_mlp": 0.01314499,
      "balance_loss_clip": 1.47196531,
      "balance_loss_mlp": 1.07112169,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.549826725868638,
      "language_loss": 0.98657227,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02202535,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.744039297103882
    },
    {
      "auxiliary_loss_clip": 0.02217803,
      "auxiliary_loss_mlp": 0.01320199,
      "balance_loss_clip": 1.46357512,
      "balance_loss_mlp": 1.06614017,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 2.2535202118579343,
      "language_loss": 0.94030154,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.97568154,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.7685961723327637
    },
    {
      "auxiliary_loss_clip": 0.02163647,
      "auxiliary_loss_mlp": 0.01300189,
      "balance_loss_clip": 1.45423543,
      "balance_loss_mlp": 1.06444025,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.6015805317765346,
      "language_loss": 0.92455447,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.95919275,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 4.7169578075408936
    },
    {
      "auxiliary_loss_clip": 0.02159091,
      "auxiliary_loss_mlp": 0.01312184,
      "balance_loss_clip": 1.44686449,
      "balance_loss_mlp": 1.06117702,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.1339366673313256,
      "language_loss": 0.92988682,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96459955,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 3.651533842086792
    },
    {
      "auxiliary_loss_clip": 0.02117032,
      "auxiliary_loss_mlp": 0.01317385,
      "balance_loss_clip": 1.43368697,
      "balance_loss_mlp": 1.08392584,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 2.5042782179467618,
      "language_loss": 0.99273968,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.02708387,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 2.8019254207611084
    },
    {
      "auxiliary_loss_clip": 0.02099079,
      "auxiliary_loss_mlp": 0.01281967,
      "balance_loss_clip": 1.43357158,
      "balance_loss_mlp": 1.05804443,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 44.22817018780025,
      "language_loss": 0.92756361,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.96137398,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 2.7972021102905273
    },
    {
      "auxiliary_loss_clip": 0.02054076,
      "auxiliary_loss_mlp": 0.01297619,
      "balance_loss_clip": 1.41339183,
      "balance_loss_mlp": 1.07369637,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.494085457534451,
      "language_loss": 0.83109963,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86461663,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.791658639907837
    },
    {
      "auxiliary_loss_clip": 0.02058007,
      "auxiliary_loss_mlp": 0.01262537,
      "balance_loss_clip": 1.41847444,
      "balance_loss_mlp": 1.05167961,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 4.112553475258077,
      "language_loss": 0.79394692,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.82715237,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.7529516220092773
    },
    {
      "auxiliary_loss_clip": 0.02037877,
      "auxiliary_loss_mlp": 0.0126015,
      "balance_loss_clip": 1.41055727,
      "balance_loss_mlp": 1.0490067,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 1.9688435939396838,
      "language_loss": 1.04321599,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07619619,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.8125345706939697
    },
    {
      "auxiliary_loss_clip": 0.02004401,
      "auxiliary_loss_mlp": 0.01253719,
      "balance_loss_clip": 1.39910614,
      "balance_loss_mlp": 1.05268407,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 2.143924458469264,
      "language_loss": 1.07885861,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11143994,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.821887969970703
    },
    {
      "auxiliary_loss_clip": 0.0198874,
      "auxiliary_loss_mlp": 0.01278063,
      "balance_loss_clip": 1.39220262,
      "balance_loss_mlp": 1.070925,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.4954135580300947,
      "language_loss": 1.08755112,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.12021911,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.7487683296203613
    },
    {
      "auxiliary_loss_clip": 0.01944578,
      "auxiliary_loss_mlp": 0.0126126,
      "balance_loss_clip": 1.38353682,
      "balance_loss_mlp": 1.0590806,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 2.9500978686394657,
      "language_loss": 1.01993632,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05199468,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.8465287685394287
    },
    {
      "auxiliary_loss_clip": 0.01928575,
      "auxiliary_loss_mlp": 0.01258553,
      "balance_loss_clip": 1.37752557,
      "balance_loss_mlp": 1.05894876,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 5.286645718614315,
      "language_loss": 0.90213507,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93400633,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.8800110816955566
    },
    {
      "auxiliary_loss_clip": 0.0192131,
      "auxiliary_loss_mlp": 0.01250717,
      "balance_loss_clip": 1.37422085,
      "balance_loss_mlp": 1.06007791,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 1.9853244312566956,
      "language_loss": 0.92470181,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.95642203,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.7842273712158203
    },
    {
      "auxiliary_loss_clip": 0.01906633,
      "auxiliary_loss_mlp": 0.01257898,
      "balance_loss_clip": 1.36807382,
      "balance_loss_mlp": 1.06115532,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.2055042664574125,
      "language_loss": 0.97950733,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.01115251,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.8949031829833984
    },
    {
      "auxiliary_loss_clip": 0.01900364,
      "auxiliary_loss_mlp": 0.01236066,
      "balance_loss_clip": 1.36323476,
      "balance_loss_mlp": 1.05057669,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 3.456860771672256,
      "language_loss": 0.88892442,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92028868,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.8582842350006104
    },
    {
      "auxiliary_loss_clip": 0.01884389,
      "auxiliary_loss_mlp": 0.0124842,
      "balance_loss_clip": 1.3572824,
      "balance_loss_mlp": 1.0647428,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 2.3206723226459367,
      "language_loss": 0.92525929,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.95658731,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.791677713394165
    },
    {
      "auxiliary_loss_clip": 0.01866479,
      "auxiliary_loss_mlp": 0.01244118,
      "balance_loss_clip": 1.35105157,
      "balance_loss_mlp": 1.0662576,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 2.458728816214611,
      "language_loss": 0.84757996,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.87868583,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.78120756149292
    },
    {
      "auxiliary_loss_clip": 0.01842597,
      "auxiliary_loss_mlp": 0.01248503,
      "balance_loss_clip": 1.34197474,
      "balance_loss_mlp": 1.0684495,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 2.1095494148805853,
      "language_loss": 0.91581672,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.94672775,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.818150281906128
    },
    {
      "auxiliary_loss_clip": 0.01828797,
      "auxiliary_loss_mlp": 0.01234313,
      "balance_loss_clip": 1.33865118,
      "balance_loss_mlp": 1.06751513,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.436825366011175,
      "language_loss": 0.86047256,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.89110374,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.7382733821868896
    },
    {
      "auxiliary_loss_clip": 0.01832322,
      "auxiliary_loss_mlp": 0.01238589,
      "balance_loss_clip": 1.33819222,
      "balance_loss_mlp": 1.06778634,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 3.074424301885012,
      "language_loss": 1.01875138,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.04946041,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.864509344100952
    },
    {
      "auxiliary_loss_clip": 0.01813821,
      "auxiliary_loss_mlp": 0.01219703,
      "balance_loss_clip": 1.33228135,
      "balance_loss_mlp": 1.05185616,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 1.881196730318686,
      "language_loss": 0.93571734,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96605265,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.7973525524139404
    },
    {
      "auxiliary_loss_clip": 0.01795228,
      "auxiliary_loss_mlp": 0.01212906,
      "balance_loss_clip": 1.32497144,
      "balance_loss_mlp": 1.04992282,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.071720224954941,
      "language_loss": 0.9737581,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00383937,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.8166849613189697
    },
    {
      "auxiliary_loss_clip": 0.01782965,
      "auxiliary_loss_mlp": 0.01214397,
      "balance_loss_clip": 1.31882048,
      "balance_loss_mlp": 1.05627811,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 2.144466701562681,
      "language_loss": 0.94086242,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97083616,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.8112339973449707
    },
    {
      "auxiliary_loss_clip": 0.01763091,
      "auxiliary_loss_mlp": 0.01214947,
      "balance_loss_clip": 1.31111288,
      "balance_loss_mlp": 1.05778122,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 2.3500046491558866,
      "language_loss": 0.72270763,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75248802,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.782559394836426
    },
    {
      "auxiliary_loss_clip": 0.0175659,
      "auxiliary_loss_mlp": 0.01212894,
      "balance_loss_clip": 1.307459,
      "balance_loss_mlp": 1.0596385,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 3.4547535184072093,
      "language_loss": 0.87950653,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.90920138,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.8497064113616943
    },
    {
      "auxiliary_loss_clip": 0.01739251,
      "auxiliary_loss_mlp": 0.01211677,
      "balance_loss_clip": 1.30153894,
      "balance_loss_mlp": 1.06643236,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 4.573660568404091,
      "language_loss": 0.96254265,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99205196,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.923413038253784
    },
    {
      "auxiliary_loss_clip": 0.01747061,
      "auxiliary_loss_mlp": 0.01204604,
      "balance_loss_clip": 1.30580282,
      "balance_loss_mlp": 1.06002736,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 2.5379260630849485,
      "language_loss": 0.88171202,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91122866,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.7812423706054688
    },
    {
      "auxiliary_loss_clip": 0.01725294,
      "auxiliary_loss_mlp": 0.01191837,
      "balance_loss_clip": 1.28956234,
      "balance_loss_mlp": 1.04878557,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.4429313660620227,
      "language_loss": 0.87840319,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.90757447,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 3.7623276710510254
    },
    {
      "auxiliary_loss_clip": 0.01710471,
      "auxiliary_loss_mlp": 0.01188176,
      "balance_loss_clip": 1.28688955,
      "balance_loss_mlp": 1.04302633,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 2.2425285114499616,
      "language_loss": 0.88801092,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.91699737,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 3.666506290435791
    },
    {
      "auxiliary_loss_clip": 0.01696097,
      "auxiliary_loss_mlp": 0.01195827,
      "balance_loss_clip": 1.28116369,
      "balance_loss_mlp": 1.05597019,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.5295616822182247,
      "language_loss": 1.04809582,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.07701504,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 2.73976469039917
    },
    {
      "auxiliary_loss_clip": 0.01687155,
      "auxiliary_loss_mlp": 0.01193415,
      "balance_loss_clip": 1.2736491,
      "balance_loss_mlp": 1.0573256,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.219357142562346,
      "language_loss": 0.95204049,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98084623,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 2.7454895973205566
    },
    {
      "auxiliary_loss_clip": 0.01603786,
      "auxiliary_loss_mlp": 0.01089123,
      "balance_loss_clip": 1.28835285,
      "balance_loss_mlp": 1.01282871,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.2998267514403847,
      "language_loss": 0.62102962,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.6479587,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 3.3891353607177734
    },
    {
      "auxiliary_loss_clip": 0.01671173,
      "auxiliary_loss_mlp": 0.01194776,
      "balance_loss_clip": 1.26893854,
      "balance_loss_mlp": 1.05639744,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 3.1859319363237955,
      "language_loss": 0.90419269,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93285221,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.7704780101776123
    },
    {
      "auxiliary_loss_clip": 0.01574643,
      "auxiliary_loss_mlp": 0.01084837,
      "balance_loss_clip": 1.27244866,
      "balance_loss_mlp": 1.01273942,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.1588247442291455,
      "language_loss": 0.58638275,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.6129775,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.208491325378418
    },
    {
      "auxiliary_loss_clip": 0.01650787,
      "auxiliary_loss_mlp": 0.01170438,
      "balance_loss_clip": 1.25353026,
      "balance_loss_mlp": 1.0506562,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.9012850472271796,
      "language_loss": 0.98896146,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.01717365,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.810664653778076
    },
    {
      "auxiliary_loss_clip": 0.01632324,
      "auxiliary_loss_mlp": 0.011795,
      "balance_loss_clip": 1.25116515,
      "balance_loss_mlp": 1.05094445,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 3.3007821427273134,
      "language_loss": 0.96822041,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99633861,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.788667917251587
    },
    {
      "auxiliary_loss_clip": 0.016286,
      "auxiliary_loss_mlp": 0.01184625,
      "balance_loss_clip": 1.248492,
      "balance_loss_mlp": 1.06017041,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 3.0346751952392395,
      "language_loss": 0.79887027,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.82700247,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.7972617149353027
    },
    {
      "auxiliary_loss_clip": 0.01611588,
      "auxiliary_loss_mlp": 0.01180862,
      "balance_loss_clip": 1.23923969,
      "balance_loss_mlp": 1.06003118,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 2.3343997057754513,
      "language_loss": 0.95568132,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98360586,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.7947232723236084
    },
    {
      "auxiliary_loss_clip": 0.01616521,
      "auxiliary_loss_mlp": 0.01171047,
      "balance_loss_clip": 1.23916459,
      "balance_loss_mlp": 1.05622482,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 2.6381353053925216,
      "language_loss": 0.79500711,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82288283,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.8492722511291504
    },
    {
      "auxiliary_loss_clip": 0.01608471,
      "auxiliary_loss_mlp": 0.01177554,
      "balance_loss_clip": 1.23278522,
      "balance_loss_mlp": 1.06625974,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.6468468105961875,
      "language_loss": 0.91596496,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94382513,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.8587656021118164
    },
    {
      "auxiliary_loss_clip": 0.01604942,
      "auxiliary_loss_mlp": 0.01161271,
      "balance_loss_clip": 1.23717332,
      "balance_loss_mlp": 1.04597163,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 4.220594998264973,
      "language_loss": 0.74977762,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77743971,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.8365020751953125
    },
    {
      "auxiliary_loss_clip": 0.0158519,
      "auxiliary_loss_mlp": 0.0116236,
      "balance_loss_clip": 1.22640848,
      "balance_loss_mlp": 1.05531037,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 2.738978791848827,
      "language_loss": 0.7606588,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.78813428,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.8469934463500977
    },
    {
      "auxiliary_loss_clip": 0.01580241,
      "auxiliary_loss_mlp": 0.01153043,
      "balance_loss_clip": 1.22663128,
      "balance_loss_mlp": 1.04666078,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 2.5658920464748065,
      "language_loss": 0.86977172,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.89710456,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.810621738433838
    },
    {
      "auxiliary_loss_clip": 0.01564569,
      "auxiliary_loss_mlp": 0.0116794,
      "balance_loss_clip": 1.21554518,
      "balance_loss_mlp": 1.05340397,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.290767903041341,
      "language_loss": 0.8935039,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92082894,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.809096097946167
    },
    {
      "auxiliary_loss_clip": 0.01567573,
      "auxiliary_loss_mlp": 0.0117184,
      "balance_loss_clip": 1.22087216,
      "balance_loss_mlp": 1.07075071,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.394775241067213,
      "language_loss": 0.83343977,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86083388,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.7184343338012695
    },
    {
      "auxiliary_loss_clip": 0.01564822,
      "auxiliary_loss_mlp": 0.01167801,
      "balance_loss_clip": 1.21933949,
      "balance_loss_mlp": 1.06470919,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 4.757095289861437,
      "language_loss": 0.95105213,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.97837836,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.7360031604766846
    },
    {
      "auxiliary_loss_clip": 0.01562163,
      "auxiliary_loss_mlp": 0.01159864,
      "balance_loss_clip": 1.21468508,
      "balance_loss_mlp": 1.05891752,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 2.0004180767270396,
      "language_loss": 0.89428413,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92150438,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.750227451324463
    },
    {
      "auxiliary_loss_clip": 0.01552791,
      "auxiliary_loss_mlp": 0.01161189,
      "balance_loss_clip": 1.21063185,
      "balance_loss_mlp": 1.06210256,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.3752904192283237,
      "language_loss": 0.93176895,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.9589088,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.752938985824585
    },
    {
      "auxiliary_loss_clip": 0.01546932,
      "auxiliary_loss_mlp": 0.0113992,
      "balance_loss_clip": 1.2087152,
      "balance_loss_mlp": 1.0426929,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 3.2205494392080705,
      "language_loss": 0.96444941,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99131793,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.7402877807617188
    },
    {
      "auxiliary_loss_clip": 0.01536077,
      "auxiliary_loss_mlp": 0.01156892,
      "balance_loss_clip": 1.2071774,
      "balance_loss_mlp": 1.05785251,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 1.8413796431387275,
      "language_loss": 0.81381607,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84074575,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.7854058742523193
    },
    {
      "auxiliary_loss_clip": 0.01535628,
      "auxiliary_loss_mlp": 0.01148931,
      "balance_loss_clip": 1.20549226,
      "balance_loss_mlp": 1.05976248,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 1.9772457907390564,
      "language_loss": 0.94058704,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96743262,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.7509875297546387
    },
    {
      "auxiliary_loss_clip": 0.01530298,
      "auxiliary_loss_mlp": 0.01135542,
      "balance_loss_clip": 1.19939089,
      "balance_loss_mlp": 1.04103315,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 3.91896009919292,
      "language_loss": 0.7520929,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.77875137,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.7212650775909424
    },
    {
      "auxiliary_loss_clip": 0.01537872,
      "auxiliary_loss_mlp": 0.01153044,
      "balance_loss_clip": 1.206424,
      "balance_loss_mlp": 1.05843949,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 2.4123536814294324,
      "language_loss": 0.84973824,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87664735,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.817582130432129
    },
    {
      "auxiliary_loss_clip": 0.01522677,
      "auxiliary_loss_mlp": 0.01144505,
      "balance_loss_clip": 1.19868875,
      "balance_loss_mlp": 1.05495501,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.6812312559575984,
      "language_loss": 0.92266858,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.9493404,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 3.922977924346924
    },
    {
      "auxiliary_loss_clip": 0.01515726,
      "auxiliary_loss_mlp": 0.01140649,
      "balance_loss_clip": 1.19271111,
      "balance_loss_mlp": 1.04799986,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 1.994407379473281,
      "language_loss": 0.96168458,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.98824835,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 4.59586501121521
    },
    {
      "auxiliary_loss_clip": 0.01523992,
      "auxiliary_loss_mlp": 0.0113666,
      "balance_loss_clip": 1.19697511,
      "balance_loss_mlp": 1.05764842,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 2.8427162385353704,
      "language_loss": 0.85247374,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.87908024,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 2.7976150512695312
    },
    {
      "auxiliary_loss_clip": 0.0151256,
      "auxiliary_loss_mlp": 0.01135555,
      "balance_loss_clip": 1.19105339,
      "balance_loss_mlp": 1.05368245,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 2.0647788360735126,
      "language_loss": 0.88761842,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91409957,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.781954765319824
    },
    {
      "auxiliary_loss_clip": 0.01514221,
      "auxiliary_loss_mlp": 0.01132428,
      "balance_loss_clip": 1.19299376,
      "balance_loss_mlp": 1.04821897,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 1.7587387437828401,
      "language_loss": 0.80381393,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.83028042,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.909999132156372
    },
    {
      "auxiliary_loss_clip": 0.01497582,
      "auxiliary_loss_mlp": 0.01135382,
      "balance_loss_clip": 1.1856389,
      "balance_loss_mlp": 1.05484462,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 2.3198670890628694,
      "language_loss": 0.92506838,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95139807,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 2.7998838424682617
    },
    {
      "auxiliary_loss_clip": 0.01500404,
      "auxiliary_loss_mlp": 0.01129406,
      "balance_loss_clip": 1.18116665,
      "balance_loss_mlp": 1.05072796,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 3.0169711288427066,
      "language_loss": 0.86680913,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89310718,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.72688889503479
    },
    {
      "auxiliary_loss_clip": 0.01495115,
      "auxiliary_loss_mlp": 0.01125006,
      "balance_loss_clip": 1.18330252,
      "balance_loss_mlp": 1.04995203,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 2.17587826712167,
      "language_loss": 1.01564705,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04184818,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.7411887645721436
    },
    {
      "auxiliary_loss_clip": 0.01493462,
      "auxiliary_loss_mlp": 0.01134296,
      "balance_loss_clip": 1.18038177,
      "balance_loss_mlp": 1.06038642,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 2.0166192151190323,
      "language_loss": 0.97754025,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.0038178,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.779336929321289
    },
    {
      "auxiliary_loss_clip": 0.01483579,
      "auxiliary_loss_mlp": 0.01135508,
      "balance_loss_clip": 1.178069,
      "balance_loss_mlp": 1.06469786,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 2.8707536561908156,
      "language_loss": 0.93887067,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96506155,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.777034282684326
    },
    {
      "auxiliary_loss_clip": 0.01484395,
      "auxiliary_loss_mlp": 0.0112259,
      "balance_loss_clip": 1.17845297,
      "balance_loss_mlp": 1.05197048,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 6.875835543669841,
      "language_loss": 0.81102765,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83709753,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.734243869781494
    },
    {
      "auxiliary_loss_clip": 0.01484316,
      "auxiliary_loss_mlp": 0.01117176,
      "balance_loss_clip": 1.17953575,
      "balance_loss_mlp": 1.04660368,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 3.389540487078456,
      "language_loss": 0.87674868,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90276366,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.6559481620788574
    },
    {
      "auxiliary_loss_clip": 0.01479638,
      "auxiliary_loss_mlp": 0.0112939,
      "balance_loss_clip": 1.17575514,
      "balance_loss_mlp": 1.06015313,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 1.9282880503669313,
      "language_loss": 0.88130212,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.90739238,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.938173532485962
    },
    {
      "auxiliary_loss_clip": 0.01477257,
      "auxiliary_loss_mlp": 0.01124444,
      "balance_loss_clip": 1.17578983,
      "balance_loss_mlp": 1.05382419,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 1.650231066134,
      "language_loss": 0.8561911,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88220811,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.8657491207122803
    },
    {
      "auxiliary_loss_clip": 0.01477366,
      "auxiliary_loss_mlp": 0.01127403,
      "balance_loss_clip": 1.1710912,
      "balance_loss_mlp": 1.06298208,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 2.8613006690534406,
      "language_loss": 0.90162194,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92766958,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.7867159843444824
    },
    {
      "auxiliary_loss_clip": 0.01475541,
      "auxiliary_loss_mlp": 0.01100802,
      "balance_loss_clip": 1.1693182,
      "balance_loss_mlp": 1.03499889,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 1.9691508351709524,
      "language_loss": 0.88480306,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91056645,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.9053361415863037
    },
    {
      "auxiliary_loss_clip": 0.01476977,
      "auxiliary_loss_mlp": 0.01116414,
      "balance_loss_clip": 1.17029607,
      "balance_loss_mlp": 1.05576003,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 2.9034418336669003,
      "language_loss": 0.96280235,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98873615,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.7983059883117676
    },
    {
      "auxiliary_loss_clip": 0.0146128,
      "auxiliary_loss_mlp": 0.01113405,
      "balance_loss_clip": 1.16531038,
      "balance_loss_mlp": 1.04669547,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 3.4401852475014953,
      "language_loss": 0.86709547,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.89284235,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.790989398956299
    },
    {
      "auxiliary_loss_clip": 0.01456177,
      "auxiliary_loss_mlp": 0.01114555,
      "balance_loss_clip": 1.16454327,
      "balance_loss_mlp": 1.0552845,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 1.9013836174535417,
      "language_loss": 0.95398986,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.97969717,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.7106270790100098
    },
    {
      "auxiliary_loss_clip": 0.01453704,
      "auxiliary_loss_mlp": 0.01115673,
      "balance_loss_clip": 1.16062534,
      "balance_loss_mlp": 1.05716538,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 1.752983957042692,
      "language_loss": 0.8112185,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.83691227,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.7654178142547607
    },
    {
      "auxiliary_loss_clip": 0.01462483,
      "auxiliary_loss_mlp": 0.01110887,
      "balance_loss_clip": 1.16443121,
      "balance_loss_mlp": 1.05037665,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 2.23689009784611,
      "language_loss": 0.76015806,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78589171,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.7869327068328857
    },
    {
      "auxiliary_loss_clip": 0.01459778,
      "auxiliary_loss_mlp": 0.01110569,
      "balance_loss_clip": 1.16549063,
      "balance_loss_mlp": 1.05134535,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 3.3184543885205495,
      "language_loss": 0.95210057,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97780406,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.7201180458068848
    },
    {
      "auxiliary_loss_clip": 0.01458532,
      "auxiliary_loss_mlp": 0.01110163,
      "balance_loss_clip": 1.16163325,
      "balance_loss_mlp": 1.05268002,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 2.0413813015077875,
      "language_loss": 0.88817143,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91385841,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.747728109359741
    },
    {
      "auxiliary_loss_clip": 0.01456742,
      "auxiliary_loss_mlp": 0.01112716,
      "balance_loss_clip": 1.16523695,
      "balance_loss_mlp": 1.05897617,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 2.475430799183803,
      "language_loss": 0.77576876,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80146337,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.8845136165618896
    },
    {
      "auxiliary_loss_clip": 0.01450588,
      "auxiliary_loss_mlp": 0.01098063,
      "balance_loss_clip": 1.15495992,
      "balance_loss_mlp": 1.03745711,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.0657318616582647,
      "language_loss": 0.9096241,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93511063,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.9041669368743896
    },
    {
      "auxiliary_loss_clip": 0.01449576,
      "auxiliary_loss_mlp": 0.01105417,
      "balance_loss_clip": 1.15606666,
      "balance_loss_mlp": 1.05096221,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 2.0340544916118772,
      "language_loss": 0.86261851,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.88816845,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 2.70540189743042
    },
    {
      "auxiliary_loss_clip": 0.01442429,
      "auxiliary_loss_mlp": 0.01104463,
      "balance_loss_clip": 1.15424657,
      "balance_loss_mlp": 1.0472424,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.9024540572394217,
      "language_loss": 0.89168489,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.91715384,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 3.765204429626465
    },
    {
      "auxiliary_loss_clip": 0.01437781,
      "auxiliary_loss_mlp": 0.01116023,
      "balance_loss_clip": 1.1523478,
      "balance_loss_mlp": 1.06459594,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 1.806766071049921,
      "language_loss": 0.92408401,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.94962204,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 4.637160301208496
    },
    {
      "auxiliary_loss_clip": 0.01448032,
      "auxiliary_loss_mlp": 0.01108621,
      "balance_loss_clip": 1.15620947,
      "balance_loss_mlp": 1.05011308,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.5269882520840152,
      "language_loss": 0.85467988,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.8802464,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 3.706763505935669
    },
    {
      "auxiliary_loss_clip": 0.01435106,
      "auxiliary_loss_mlp": 0.01098133,
      "balance_loss_clip": 1.14920521,
      "balance_loss_mlp": 1.04451287,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 2.0824040084400512,
      "language_loss": 0.83126497,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85659736,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.8266806602478027
    },
    {
      "auxiliary_loss_clip": 0.01434528,
      "auxiliary_loss_mlp": 0.01100971,
      "balance_loss_clip": 1.14912331,
      "balance_loss_mlp": 1.04816127,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 2.1234942193359525,
      "language_loss": 0.88723534,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91259038,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.798060655593872
    },
    {
      "auxiliary_loss_clip": 0.01436204,
      "auxiliary_loss_mlp": 0.01106911,
      "balance_loss_clip": 1.14992511,
      "balance_loss_mlp": 1.05691409,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 2.1117918111629135,
      "language_loss": 0.89790356,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.92333472,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.7501001358032227
    },
    {
      "auxiliary_loss_clip": 0.01429196,
      "auxiliary_loss_mlp": 0.01094674,
      "balance_loss_clip": 1.14724386,
      "balance_loss_mlp": 1.04782438,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 2.0388835135399406,
      "language_loss": 0.83994025,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86517888,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 2.787310838699341
    },
    {
      "auxiliary_loss_clip": 0.01426915,
      "auxiliary_loss_mlp": 0.01102147,
      "balance_loss_clip": 1.14474726,
      "balance_loss_mlp": 1.05293703,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 2.4358841681042187,
      "language_loss": 0.86811411,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89340466,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 2.6752512454986572
    },
    {
      "auxiliary_loss_clip": 0.01428633,
      "auxiliary_loss_mlp": 0.01102452,
      "balance_loss_clip": 1.14464772,
      "balance_loss_mlp": 1.05374336,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 2.1588874630789427,
      "language_loss": 0.968225,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.99353576,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.7756388187408447
    },
    {
      "auxiliary_loss_clip": 0.01427342,
      "auxiliary_loss_mlp": 0.01088421,
      "balance_loss_clip": 1.14310789,
      "balance_loss_mlp": 1.04166698,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 2.3217971322988915,
      "language_loss": 0.8251406,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85029817,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.8891191482543945
    },
    {
      "auxiliary_loss_clip": 0.01423095,
      "auxiliary_loss_mlp": 0.01088894,
      "balance_loss_clip": 1.14165354,
      "balance_loss_mlp": 1.04185367,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.0045216201446854,
      "language_loss": 0.78412199,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.80924189,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.9337754249572754
    },
    {
      "auxiliary_loss_clip": 0.01418247,
      "auxiliary_loss_mlp": 0.01096277,
      "balance_loss_clip": 1.14000511,
      "balance_loss_mlp": 1.04866505,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 1.8829458860688257,
      "language_loss": 0.8396486,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.86479378,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.7400522232055664
    },
    {
      "auxiliary_loss_clip": 0.01431873,
      "auxiliary_loss_mlp": 0.01107127,
      "balance_loss_clip": 1.15068018,
      "balance_loss_mlp": 1.05622435,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 2.936924842199411,
      "language_loss": 0.87714565,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90253562,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.8518614768981934
    },
    {
      "auxiliary_loss_clip": 0.0147705,
      "auxiliary_loss_mlp": 0.01047416,
      "balance_loss_clip": 1.24932718,
      "balance_loss_mlp": 1.0218575,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.034354854534219,
      "language_loss": 0.71520704,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.74045169,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.354323625564575
    },
    {
      "auxiliary_loss_clip": 0.01414398,
      "auxiliary_loss_mlp": 0.01108343,
      "balance_loss_clip": 1.13749123,
      "balance_loss_mlp": 1.06044436,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 3.8316805273671926,
      "language_loss": 0.8226009,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.84782833,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.8335580825805664
    },
    {
      "auxiliary_loss_clip": 0.01410484,
      "auxiliary_loss_mlp": 0.01087375,
      "balance_loss_clip": 1.13413739,
      "balance_loss_mlp": 1.04596174,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 2.295363233538807,
      "language_loss": 0.90258223,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.92756081,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.8527915477752686
    },
    {
      "auxiliary_loss_clip": 0.01413996,
      "auxiliary_loss_mlp": 0.01089292,
      "balance_loss_clip": 1.1383698,
      "balance_loss_mlp": 1.0492382,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 2.240721984769087,
      "language_loss": 0.86962557,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89465839,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.7263360023498535
    },
    {
      "auxiliary_loss_clip": 0.01412893,
      "auxiliary_loss_mlp": 0.01091897,
      "balance_loss_clip": 1.13600302,
      "balance_loss_mlp": 1.04833817,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 2.798652089333793,
      "language_loss": 0.76097536,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78602326,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.8291869163513184
    },
    {
      "auxiliary_loss_clip": 0.01409765,
      "auxiliary_loss_mlp": 0.010846,
      "balance_loss_clip": 1.13389206,
      "balance_loss_mlp": 1.04287624,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 2.243809666548272,
      "language_loss": 0.83312851,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.85807216,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.7921762466430664
    },
    {
      "auxiliary_loss_clip": 0.01408099,
      "auxiliary_loss_mlp": 0.01087257,
      "balance_loss_clip": 1.13358688,
      "balance_loss_mlp": 1.04493737,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 1.6987356152795927,
      "language_loss": 0.76823217,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79318571,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.7114295959472656
    },
    {
      "auxiliary_loss_clip": 0.01414616,
      "auxiliary_loss_mlp": 0.0111384,
      "balance_loss_clip": 1.14077044,
      "balance_loss_mlp": 1.07109118,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 2.425252580650448,
      "language_loss": 0.84153837,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86682296,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.8083784580230713
    },
    {
      "auxiliary_loss_clip": 0.01407977,
      "auxiliary_loss_mlp": 0.0108317,
      "balance_loss_clip": 1.13221502,
      "balance_loss_mlp": 1.04194784,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 1.738319234509562,
      "language_loss": 0.89907324,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92398471,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.736917495727539
    },
    {
      "auxiliary_loss_clip": 0.01410843,
      "auxiliary_loss_mlp": 0.01090682,
      "balance_loss_clip": 1.13356233,
      "balance_loss_mlp": 1.0496738,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 4.329679301988367,
      "language_loss": 0.90983766,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93485302,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.707399368286133
    },
    {
      "auxiliary_loss_clip": 0.01402501,
      "auxiliary_loss_mlp": 0.01085241,
      "balance_loss_clip": 1.13133824,
      "balance_loss_mlp": 1.04370821,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 2.167392985463678,
      "language_loss": 0.88083106,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90570855,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.7193057537078857
    },
    {
      "auxiliary_loss_clip": 0.01406484,
      "auxiliary_loss_mlp": 0.01100138,
      "balance_loss_clip": 1.13137007,
      "balance_loss_mlp": 1.05467129,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 2.1219657913225816,
      "language_loss": 0.78596741,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.81103367,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.7812976837158203
    },
    {
      "auxiliary_loss_clip": 0.01397753,
      "auxiliary_loss_mlp": 0.01089602,
      "balance_loss_clip": 1.12918568,
      "balance_loss_mlp": 1.05002451,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.004976270626316,
      "language_loss": 0.87466812,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.89954174,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.6732726097106934
    },
    {
      "auxiliary_loss_clip": 0.01398209,
      "auxiliary_loss_mlp": 0.01089999,
      "balance_loss_clip": 1.1275568,
      "balance_loss_mlp": 1.0510416,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 1.8928630708794982,
      "language_loss": 0.80518627,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.83006829,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 3.703705072402954
    },
    {
      "auxiliary_loss_clip": 0.01399885,
      "auxiliary_loss_mlp": 0.0110267,
      "balance_loss_clip": 1.12803328,
      "balance_loss_mlp": 1.06421292,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 1.8741605374397714,
      "language_loss": 0.97098404,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99600953,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 3.612124443054199
    },
    {
      "auxiliary_loss_clip": 0.01404093,
      "auxiliary_loss_mlp": 0.01024416,
      "balance_loss_clip": 1.19178367,
      "balance_loss_mlp": 1.00133741,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0507977384380534,
      "language_loss": 0.58418441,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60846949,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 4.201429843902588
    },
    {
      "auxiliary_loss_clip": 0.01390371,
      "auxiliary_loss_mlp": 0.01088318,
      "balance_loss_clip": 1.12208343,
      "balance_loss_mlp": 1.0525316,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 2.508722123557612,
      "language_loss": 0.85574704,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.88053393,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 2.7530224323272705
    },
    {
      "auxiliary_loss_clip": 0.01392544,
      "auxiliary_loss_mlp": 0.01076318,
      "balance_loss_clip": 1.12440968,
      "balance_loss_mlp": 1.03788507,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.5493355184899715,
      "language_loss": 0.88804615,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91273475,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.7226905822753906
    },
    {
      "auxiliary_loss_clip": 0.01387817,
      "auxiliary_loss_mlp": 0.01075233,
      "balance_loss_clip": 1.12215137,
      "balance_loss_mlp": 1.03832603,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 2.1598483622943903,
      "language_loss": 0.88271588,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.90734637,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.702954053878784
    },
    {
      "auxiliary_loss_clip": 0.01389365,
      "auxiliary_loss_mlp": 0.01091172,
      "balance_loss_clip": 1.12154543,
      "balance_loss_mlp": 1.05507541,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 2.2077200531333347,
      "language_loss": 0.86413914,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.88894451,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.796477794647217
    },
    {
      "auxiliary_loss_clip": 0.01393626,
      "auxiliary_loss_mlp": 0.01083717,
      "balance_loss_clip": 1.12300491,
      "balance_loss_mlp": 1.04640436,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 7.05511520552158,
      "language_loss": 0.93628347,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96105689,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 2.7627451419830322
    },
    {
      "auxiliary_loss_clip": 0.0137941,
      "auxiliary_loss_mlp": 0.01087807,
      "balance_loss_clip": 1.11491776,
      "balance_loss_mlp": 1.0511626,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 1.772004659630072,
      "language_loss": 0.85623938,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.88091159,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 2.8304662704467773
    },
    {
      "auxiliary_loss_clip": 0.01383751,
      "auxiliary_loss_mlp": 0.01093014,
      "balance_loss_clip": 1.11867595,
      "balance_loss_mlp": 1.05624986,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 2.934851762314538,
      "language_loss": 0.88875711,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91352475,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 2.6817362308502197
    },
    {
      "auxiliary_loss_clip": 0.01388579,
      "auxiliary_loss_mlp": 0.01089264,
      "balance_loss_clip": 1.12014794,
      "balance_loss_mlp": 1.05216575,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 2.746939211522687,
      "language_loss": 0.9580251,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98280346,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.737531900405884
    },
    {
      "auxiliary_loss_clip": 0.01384287,
      "auxiliary_loss_mlp": 0.01078624,
      "balance_loss_clip": 1.1220423,
      "balance_loss_mlp": 1.04419684,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 2.289419950971102,
      "language_loss": 0.90474176,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.92937088,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.684894561767578
    },
    {
      "auxiliary_loss_clip": 0.01377568,
      "auxiliary_loss_mlp": 0.01075928,
      "balance_loss_clip": 1.11102676,
      "balance_loss_mlp": 1.04037929,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 2.5307630857015786,
      "language_loss": 0.76233053,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78686547,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.734891891479492
    },
    {
      "auxiliary_loss_clip": 0.0137918,
      "auxiliary_loss_mlp": 0.01065948,
      "balance_loss_clip": 1.11586201,
      "balance_loss_mlp": 1.03206861,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.4190343604320454,
      "language_loss": 1.04121923,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.06567049,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.717433452606201
    },
    {
      "auxiliary_loss_clip": 0.01381726,
      "auxiliary_loss_mlp": 0.01071216,
      "balance_loss_clip": 1.11668026,
      "balance_loss_mlp": 1.03690767,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.6122358715373037,
      "language_loss": 0.85375851,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.87828791,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.6094589233398438
    },
    {
      "auxiliary_loss_clip": 0.01374536,
      "auxiliary_loss_mlp": 0.01082752,
      "balance_loss_clip": 1.10974801,
      "balance_loss_mlp": 1.05016017,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 2.4375298892428208,
      "language_loss": 0.88701355,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.9115864,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.7333104610443115
    },
    {
      "auxiliary_loss_clip": 0.0137595,
      "auxiliary_loss_mlp": 0.01077424,
      "balance_loss_clip": 1.11380601,
      "balance_loss_mlp": 1.04335451,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 2.063695412784422,
      "language_loss": 0.86271811,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.88725185,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.6537106037139893
    },
    {
      "auxiliary_loss_clip": 0.01377868,
      "auxiliary_loss_mlp": 0.0108042,
      "balance_loss_clip": 1.11676824,
      "balance_loss_mlp": 1.04801941,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 3.3820366762993093,
      "language_loss": 0.87688017,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90146303,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.783328056335449
    },
    {
      "auxiliary_loss_clip": 0.01369602,
      "auxiliary_loss_mlp": 0.01089542,
      "balance_loss_clip": 1.10907066,
      "balance_loss_mlp": 1.05509019,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 5.897061300783691,
      "language_loss": 0.81908345,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.8436749,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.8100359439849854
    },
    {
      "auxiliary_loss_clip": 0.01374243,
      "auxiliary_loss_mlp": 0.01086312,
      "balance_loss_clip": 1.11087704,
      "balance_loss_mlp": 1.05083513,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 2.4752146681662883,
      "language_loss": 0.9422214,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.96682692,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.7646701335906982
    },
    {
      "auxiliary_loss_clip": 0.01374954,
      "auxiliary_loss_mlp": 0.01086527,
      "balance_loss_clip": 1.11148202,
      "balance_loss_mlp": 1.05317235,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.2528898344093675,
      "language_loss": 0.87445205,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89906687,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.678227663040161
    },
    {
      "auxiliary_loss_clip": 0.01372843,
      "auxiliary_loss_mlp": 0.01082727,
      "balance_loss_clip": 1.11297035,
      "balance_loss_mlp": 1.04891944,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.3087565474589478,
      "language_loss": 0.73751879,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76207447,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.721855640411377
    },
    {
      "auxiliary_loss_clip": 0.01367522,
      "auxiliary_loss_mlp": 0.01077184,
      "balance_loss_clip": 1.10641289,
      "balance_loss_mlp": 1.04359126,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.1599769126087778,
      "language_loss": 0.79916817,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82361519,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.7804155349731445
    },
    {
      "auxiliary_loss_clip": 0.01367346,
      "auxiliary_loss_mlp": 0.01090152,
      "balance_loss_clip": 1.10961926,
      "balance_loss_mlp": 1.05653536,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 3.540123885987892,
      "language_loss": 0.82829487,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85286987,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.676445484161377
    },
    {
      "auxiliary_loss_clip": 0.01372166,
      "auxiliary_loss_mlp": 0.0108413,
      "balance_loss_clip": 1.11094534,
      "balance_loss_mlp": 1.05020285,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.7341441601237406,
      "language_loss": 0.97065806,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99522096,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.6942994594573975
    },
    {
      "auxiliary_loss_clip": 0.01369286,
      "auxiliary_loss_mlp": 0.01071414,
      "balance_loss_clip": 1.10721612,
      "balance_loss_mlp": 1.03863096,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 2.1298611901045876,
      "language_loss": 0.84825754,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87266445,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.8514304161071777
    },
    {
      "auxiliary_loss_clip": 0.01365952,
      "auxiliary_loss_mlp": 0.0107805,
      "balance_loss_clip": 1.10730815,
      "balance_loss_mlp": 1.04841447,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.452560502852618,
      "language_loss": 0.82848603,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85292602,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 2.8251020908355713
    },
    {
      "auxiliary_loss_clip": 0.01364928,
      "auxiliary_loss_mlp": 0.01076637,
      "balance_loss_clip": 1.10654247,
      "balance_loss_mlp": 1.04516613,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 6.815778605514142,
      "language_loss": 0.86413366,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.88854933,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 4.642110109329224
    },
    {
      "auxiliary_loss_clip": 0.01366191,
      "auxiliary_loss_mlp": 0.01077874,
      "balance_loss_clip": 1.10778582,
      "balance_loss_mlp": 1.04533017,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 2.315916625536218,
      "language_loss": 0.97485292,
      "learning_rate": 3.729081526348224e-06,
      "loss": 0.99929357,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 3.713261604309082
    },
    {
      "auxiliary_loss_clip": 0.01363319,
      "auxiliary_loss_mlp": 0.01070251,
      "balance_loss_clip": 1.10550499,
      "balance_loss_mlp": 1.03780222,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 1.765911659289122,
      "language_loss": 0.85082412,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87515986,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 2.796400308609009
    },
    {
      "auxiliary_loss_clip": 0.0136408,
      "auxiliary_loss_mlp": 0.01076739,
      "balance_loss_clip": 1.10752773,
      "balance_loss_mlp": 1.04512429,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 3.035002362356243,
      "language_loss": 0.9510386,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.9754467,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.7366597652435303
    },
    {
      "auxiliary_loss_clip": 0.01361749,
      "auxiliary_loss_mlp": 0.01072852,
      "balance_loss_clip": 1.10672081,
      "balance_loss_mlp": 1.04095197,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 1.8185495640913043,
      "language_loss": 0.90805662,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93240261,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.822354555130005
    },
    {
      "auxiliary_loss_clip": 0.01366643,
      "auxiliary_loss_mlp": 0.01085639,
      "balance_loss_clip": 1.10589266,
      "balance_loss_mlp": 1.05218852,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 2.076994759158947,
      "language_loss": 0.89300144,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.91752428,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.7950334548950195
    },
    {
      "auxiliary_loss_clip": 0.01360056,
      "auxiliary_loss_mlp": 0.01067792,
      "balance_loss_clip": 1.10451031,
      "balance_loss_mlp": 1.0365591,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 2.8585790927405994,
      "language_loss": 0.72774327,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.75202179,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.794891119003296
    },
    {
      "auxiliary_loss_clip": 0.01365424,
      "auxiliary_loss_mlp": 0.01080412,
      "balance_loss_clip": 1.10693729,
      "balance_loss_mlp": 1.04808211,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 2.948632077161472,
      "language_loss": 0.98119462,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00565302,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 2.676165819168091
    },
    {
      "auxiliary_loss_clip": 0.01362085,
      "auxiliary_loss_mlp": 0.01075453,
      "balance_loss_clip": 1.10481203,
      "balance_loss_mlp": 1.043648,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 3.946655758873742,
      "language_loss": 0.93728751,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.96166289,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 2.7103846073150635
    },
    {
      "auxiliary_loss_clip": 0.01363637,
      "auxiliary_loss_mlp": 0.01076617,
      "balance_loss_clip": 1.10566497,
      "balance_loss_mlp": 1.04497862,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.498527398970377,
      "language_loss": 0.89441907,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.91882169,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 2.71836256980896
    },
    {
      "auxiliary_loss_clip": 0.01362889,
      "auxiliary_loss_mlp": 0.01084502,
      "balance_loss_clip": 1.1052928,
      "balance_loss_mlp": 1.05474746,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 1.8360745307311057,
      "language_loss": 0.94365728,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.96813118,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.7241501808166504
    },
    {
      "auxiliary_loss_clip": 0.01353767,
      "auxiliary_loss_mlp": 0.01066868,
      "balance_loss_clip": 1.09885478,
      "balance_loss_mlp": 1.03539693,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 2.1018421892561236,
      "language_loss": 0.90228653,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.92649293,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.737093210220337
    },
    {
      "auxiliary_loss_clip": 0.013588,
      "auxiliary_loss_mlp": 0.01076393,
      "balance_loss_clip": 1.10274744,
      "balance_loss_mlp": 1.04351485,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 1.9019508538440082,
      "language_loss": 0.82120854,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84556055,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.780498743057251
    },
    {
      "auxiliary_loss_clip": 0.01354533,
      "auxiliary_loss_mlp": 0.01073144,
      "balance_loss_clip": 1.09957719,
      "balance_loss_mlp": 1.04145813,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 1.77338251694263,
      "language_loss": 0.80792928,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.83220613,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.894601821899414
    },
    {
      "auxiliary_loss_clip": 0.01358755,
      "auxiliary_loss_mlp": 0.01074548,
      "balance_loss_clip": 1.10013509,
      "balance_loss_mlp": 1.04329145,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 2.2848425747529633,
      "language_loss": 0.85840887,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88274193,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.7528247833251953
    },
    {
      "auxiliary_loss_clip": 0.01352767,
      "auxiliary_loss_mlp": 0.01083098,
      "balance_loss_clip": 1.09890413,
      "balance_loss_mlp": 1.05117369,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 2.448344914460138,
      "language_loss": 0.77065176,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.79501039,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.7822039127349854
    },
    {
      "auxiliary_loss_clip": 0.01355251,
      "auxiliary_loss_mlp": 0.01072314,
      "balance_loss_clip": 1.10182607,
      "balance_loss_mlp": 1.0417726,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 2.0044047028130825,
      "language_loss": 0.79090834,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81518394,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.971869707107544
    },
    {
      "auxiliary_loss_clip": 0.01354337,
      "auxiliary_loss_mlp": 0.01062054,
      "balance_loss_clip": 1.09999347,
      "balance_loss_mlp": 1.03284776,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 1.9822016558472235,
      "language_loss": 0.84392262,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.86808658,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.8568034172058105
    },
    {
      "auxiliary_loss_clip": 0.01349634,
      "auxiliary_loss_mlp": 0.01071481,
      "balance_loss_clip": 1.09792471,
      "balance_loss_mlp": 1.04206038,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 1.8551135978686237,
      "language_loss": 0.8872456,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.9114567,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.817115306854248
    },
    {
      "auxiliary_loss_clip": 0.01358126,
      "auxiliary_loss_mlp": 0.01070464,
      "balance_loss_clip": 1.1024332,
      "balance_loss_mlp": 1.03784811,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 2.4411419543810964,
      "language_loss": 0.78200948,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.80629539,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.6819498538970947
    },
    {
      "auxiliary_loss_clip": 0.01350054,
      "auxiliary_loss_mlp": 0.01063786,
      "balance_loss_clip": 1.09790564,
      "balance_loss_mlp": 1.0342226,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 3.9576796182376737,
      "language_loss": 0.89456159,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.91869998,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.6234421730041504
    },
    {
      "auxiliary_loss_clip": 0.01346265,
      "auxiliary_loss_mlp": 0.01019002,
      "balance_loss_clip": 1.15299368,
      "balance_loss_mlp": 0.99783033,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.019427291436444,
      "language_loss": 0.63212889,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.65578157,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.218242645263672
    },
    {
      "auxiliary_loss_clip": 0.01347568,
      "auxiliary_loss_mlp": 0.0107117,
      "balance_loss_clip": 1.09610176,
      "balance_loss_mlp": 1.0412488,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 2.2535662489363113,
      "language_loss": 0.84537089,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.86955827,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.706188678741455
    },
    {
      "auxiliary_loss_clip": 0.01350683,
      "auxiliary_loss_mlp": 0.01077079,
      "balance_loss_clip": 1.09627473,
      "balance_loss_mlp": 1.04706228,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 2.624683274630268,
      "language_loss": 0.88149589,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.90577352,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.7797164916992188
    },
    {
      "auxiliary_loss_clip": 0.01351577,
      "auxiliary_loss_mlp": 0.01085471,
      "balance_loss_clip": 1.0984205,
      "balance_loss_mlp": 1.05490601,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.1458932516656426,
      "language_loss": 0.86423361,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.88860404,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.749197006225586
    },
    {
      "auxiliary_loss_clip": 0.01351168,
      "auxiliary_loss_mlp": 0.01080445,
      "balance_loss_clip": 1.09838223,
      "balance_loss_mlp": 1.05202508,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.3997893883473345,
      "language_loss": 0.82537609,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.84969223,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.6551928520202637
    },
    {
      "auxiliary_loss_clip": 0.01349747,
      "auxiliary_loss_mlp": 0.01085047,
      "balance_loss_clip": 1.09975743,
      "balance_loss_mlp": 1.05557847,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 5.11499363813138,
      "language_loss": 0.84685999,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.87120795,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 3.6538140773773193
    },
    {
      "auxiliary_loss_clip": 0.01350484,
      "auxiliary_loss_mlp": 0.01081758,
      "balance_loss_clip": 1.09930587,
      "balance_loss_mlp": 1.05076361,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 3.5902767270629443,
      "language_loss": 0.78814387,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81246626,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 3.5782968997955322
    },
    {
      "auxiliary_loss_clip": 0.0135117,
      "auxiliary_loss_mlp": 0.01072876,
      "balance_loss_clip": 1.10516047,
      "balance_loss_mlp": 1.04362202,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 2.737865105476419,
      "language_loss": 0.76051271,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.7847532,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 4.422708988189697
    },
    {
      "auxiliary_loss_clip": 0.01348586,
      "auxiliary_loss_mlp": 0.01064193,
      "balance_loss_clip": 1.09678233,
      "balance_loss_mlp": 1.03462863,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 2.564441828849806,
      "language_loss": 0.87167037,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89579821,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 2.6950788497924805
    },
    {
      "auxiliary_loss_clip": 0.01343749,
      "auxiliary_loss_mlp": 0.01070193,
      "balance_loss_clip": 1.09375393,
      "balance_loss_mlp": 1.04250062,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 2.778046505885696,
      "language_loss": 0.84907341,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.87321281,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.6996614933013916
    },
    {
      "auxiliary_loss_clip": 0.01343085,
      "auxiliary_loss_mlp": 0.01064802,
      "balance_loss_clip": 1.09474158,
      "balance_loss_mlp": 1.03623962,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 1.947605264421667,
      "language_loss": 0.87892032,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.90299922,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.7209503650665283
    },
    {
      "auxiliary_loss_clip": 0.01345884,
      "auxiliary_loss_mlp": 0.01086517,
      "balance_loss_clip": 1.10021853,
      "balance_loss_mlp": 1.05771613,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 1.8302845108892802,
      "language_loss": 0.77365136,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.7979753,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.7838692665100098
    },
    {
      "auxiliary_loss_clip": 0.01344457,
      "auxiliary_loss_mlp": 0.01076582,
      "balance_loss_clip": 1.09541512,
      "balance_loss_mlp": 1.04627943,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 1.7449039585086255,
      "language_loss": 0.83758575,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.86179614,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 2.7639167308807373
    },
    {
      "auxiliary_loss_clip": 0.01341731,
      "auxiliary_loss_mlp": 0.0106187,
      "balance_loss_clip": 1.09302378,
      "balance_loss_mlp": 1.03237748,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 2.3892280505199865,
      "language_loss": 0.91007835,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93411434,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 2.725076198577881
    },
    {
      "auxiliary_loss_clip": 0.0133654,
      "auxiliary_loss_mlp": 0.01067566,
      "balance_loss_clip": 1.09148061,
      "balance_loss_mlp": 1.04091108,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.004810950065198,
      "language_loss": 0.91282564,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.93686664,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 2.7983551025390625
    },
    {
      "auxiliary_loss_clip": 0.01345279,
      "auxiliary_loss_mlp": 0.01074397,
      "balance_loss_clip": 1.09826636,
      "balance_loss_mlp": 1.04590607,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 2.2738405940201214,
      "language_loss": 0.9177134,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.94191021,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.744093179702759
    },
    {
      "auxiliary_loss_clip": 0.01339222,
      "auxiliary_loss_mlp": 0.0107511,
      "balance_loss_clip": 1.09389687,
      "balance_loss_mlp": 1.04737043,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 2.265971898717229,
      "language_loss": 0.8593263,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88346964,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.7320077419281006
    },
    {
      "auxiliary_loss_clip": 0.01344731,
      "auxiliary_loss_mlp": 0.01069079,
      "balance_loss_clip": 1.09651554,
      "balance_loss_mlp": 1.04051626,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 3.2690478286694735,
      "language_loss": 0.91356826,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.93770635,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.724618911743164
    },
    {
      "auxiliary_loss_clip": 0.01340495,
      "auxiliary_loss_mlp": 0.01070072,
      "balance_loss_clip": 1.09234452,
      "balance_loss_mlp": 1.04267752,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.094663578833228,
      "language_loss": 0.92316067,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.94726634,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.6325855255126953
    },
    {
      "auxiliary_loss_clip": 0.01337738,
      "auxiliary_loss_mlp": 0.01069722,
      "balance_loss_clip": 1.09267938,
      "balance_loss_mlp": 1.04049158,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 2.6921732077599403,
      "language_loss": 0.90955412,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93362868,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.655782699584961
    },
    {
      "auxiliary_loss_clip": 0.01339822,
      "auxiliary_loss_mlp": 0.01083525,
      "balance_loss_clip": 1.09241438,
      "balance_loss_mlp": 1.05520034,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 2.112388084054683,
      "language_loss": 0.78337145,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80760491,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.7007598876953125
    },
    {
      "auxiliary_loss_clip": 0.01337161,
      "auxiliary_loss_mlp": 0.01065349,
      "balance_loss_clip": 1.09019554,
      "balance_loss_mlp": 1.03690577,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 1.9050537163008023,
      "language_loss": 0.87135315,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.89537829,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.6519830226898193
    },
    {
      "auxiliary_loss_clip": 0.01341428,
      "auxiliary_loss_mlp": 0.01066859,
      "balance_loss_clip": 1.09598637,
      "balance_loss_mlp": 1.03965557,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 1.9216226294464447,
      "language_loss": 0.89815706,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.9222399,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.6289045810699463
    },
    {
      "auxiliary_loss_clip": 0.01334123,
      "auxiliary_loss_mlp": 0.01064258,
      "balance_loss_clip": 1.09235036,
      "balance_loss_mlp": 1.0365541,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 1.6241758902454755,
      "language_loss": 0.92227036,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94625419,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.6398439407348633
    },
    {
      "auxiliary_loss_clip": 0.01337411,
      "auxiliary_loss_mlp": 0.01065299,
      "balance_loss_clip": 1.09199119,
      "balance_loss_mlp": 1.03848851,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 2.3495783421615157,
      "language_loss": 0.85227561,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87630272,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.686149835586548
    },
    {
      "auxiliary_loss_clip": 0.01334611,
      "auxiliary_loss_mlp": 0.01068485,
      "balance_loss_clip": 1.09259605,
      "balance_loss_mlp": 1.04049468,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 1.773332424382651,
      "language_loss": 0.75175822,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.7757892,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.7387919425964355
    },
    {
      "auxiliary_loss_clip": 0.01340724,
      "auxiliary_loss_mlp": 0.01069922,
      "balance_loss_clip": 1.09379494,
      "balance_loss_mlp": 1.04216969,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 2.3623934364442216,
      "language_loss": 0.87454277,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89864922,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.6601178646087646
    },
    {
      "auxiliary_loss_clip": 0.01345291,
      "auxiliary_loss_mlp": 0.01059097,
      "balance_loss_clip": 1.09750843,
      "balance_loss_mlp": 1.03096366,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 2.9871486806054515,
      "language_loss": 0.74316996,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.76721382,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.852297067642212
    },
    {
      "auxiliary_loss_clip": 0.01335723,
      "auxiliary_loss_mlp": 0.01074941,
      "balance_loss_clip": 1.09376311,
      "balance_loss_mlp": 1.04757023,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.039110752351116,
      "language_loss": 0.8189019,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84300852,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.6257381439208984
    },
    {
      "auxiliary_loss_clip": 0.01334686,
      "auxiliary_loss_mlp": 0.01062534,
      "balance_loss_clip": 1.09210205,
      "balance_loss_mlp": 1.03525877,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.3646098469953634,
      "language_loss": 0.88485396,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90882611,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.631923198699951
    },
    {
      "auxiliary_loss_clip": 0.01335225,
      "auxiliary_loss_mlp": 0.01068749,
      "balance_loss_clip": 1.09215641,
      "balance_loss_mlp": 1.0421772,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.495656178816696,
      "language_loss": 0.84422481,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86826456,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.705756664276123
    },
    {
      "auxiliary_loss_clip": 0.01336869,
      "auxiliary_loss_mlp": 0.0107488,
      "balance_loss_clip": 1.09448564,
      "balance_loss_mlp": 1.04693711,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 2.682340461183856,
      "language_loss": 0.76596588,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.79008335,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 2.646345853805542
    },
    {
      "auxiliary_loss_clip": 0.01336172,
      "auxiliary_loss_mlp": 0.01066871,
      "balance_loss_clip": 1.09050369,
      "balance_loss_mlp": 1.03878498,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 1.7035982712761666,
      "language_loss": 0.94988763,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97391808,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 3.5990734100341797
    },
    {
      "auxiliary_loss_clip": 0.01334582,
      "auxiliary_loss_mlp": 0.01073995,
      "balance_loss_clip": 1.09163952,
      "balance_loss_mlp": 1.04745924,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 2.4969445658399243,
      "language_loss": 0.79693806,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82102376,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 4.600550651550293
    },
    {
      "auxiliary_loss_clip": 0.01332416,
      "auxiliary_loss_mlp": 0.01072312,
      "balance_loss_clip": 1.09160972,
      "balance_loss_mlp": 1.04575181,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 1.9575204741393715,
      "language_loss": 0.84118903,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86523628,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 2.770097017288208
    },
    {
      "auxiliary_loss_clip": 0.01332433,
      "auxiliary_loss_mlp": 0.01063276,
      "balance_loss_clip": 1.09111071,
      "balance_loss_mlp": 1.03675246,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 2.7766356357608726,
      "language_loss": 0.77470219,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79865932,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 2.748331308364868
    },
    {
      "auxiliary_loss_clip": 0.01318413,
      "auxiliary_loss_mlp": 0.01051288,
      "balance_loss_clip": 1.12957823,
      "balance_loss_mlp": 1.02916241,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 0.9833387857261546,
      "language_loss": 0.59288347,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61658049,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.268104314804077
    },
    {
      "auxiliary_loss_clip": 0.01336578,
      "auxiliary_loss_mlp": 0.01067695,
      "balance_loss_clip": 1.09661722,
      "balance_loss_mlp": 1.04108763,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 1.9192401743598655,
      "language_loss": 0.84389657,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86793923,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.743434190750122
    },
    {
      "auxiliary_loss_clip": 0.01333768,
      "auxiliary_loss_mlp": 0.01064506,
      "balance_loss_clip": 1.09338701,
      "balance_loss_mlp": 1.03754139,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 2.1200375194018783,
      "language_loss": 0.79598153,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.81996429,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.770836591720581
    },
    {
      "auxiliary_loss_clip": 0.01334586,
      "auxiliary_loss_mlp": 0.01078884,
      "balance_loss_clip": 1.09077573,
      "balance_loss_mlp": 1.05233645,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 2.319534973028515,
      "language_loss": 0.9057318,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.92986649,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.695350170135498
    },
    {
      "auxiliary_loss_clip": 0.01333755,
      "auxiliary_loss_mlp": 0.01071099,
      "balance_loss_clip": 1.09413362,
      "balance_loss_mlp": 1.04499173,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 2.041666861544963,
      "language_loss": 0.85850513,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.8825537,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 2.9172284603118896
    },
    {
      "auxiliary_loss_clip": 0.01331233,
      "auxiliary_loss_mlp": 0.01068574,
      "balance_loss_clip": 1.08807802,
      "balance_loss_mlp": 1.04157329,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 4.920655946657788,
      "language_loss": 0.86658221,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.8905803,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 2.753263473510742
    },
    {
      "auxiliary_loss_clip": 0.01332865,
      "auxiliary_loss_mlp": 0.01067236,
      "balance_loss_clip": 1.09086514,
      "balance_loss_mlp": 1.0399605,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.258860824789818,
      "language_loss": 0.82080585,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84480679,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 2.633981466293335
    },
    {
      "auxiliary_loss_clip": 0.01331365,
      "auxiliary_loss_mlp": 0.01067306,
      "balance_loss_clip": 1.09400249,
      "balance_loss_mlp": 1.04045975,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.081354761983096,
      "language_loss": 0.81907266,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84305936,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.746469259262085
    },
    {
      "auxiliary_loss_clip": 0.01333942,
      "auxiliary_loss_mlp": 0.01069878,
      "balance_loss_clip": 1.09352171,
      "balance_loss_mlp": 1.04257917,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 2.0717758370152692,
      "language_loss": 0.84328347,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86732173,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.665515422821045
    },
    {
      "auxiliary_loss_clip": 0.01333717,
      "auxiliary_loss_mlp": 0.01068421,
      "balance_loss_clip": 1.08935833,
      "balance_loss_mlp": 1.04176617,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 3.0120691442903182,
      "language_loss": 0.94016111,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.9641825,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.6792736053466797
    },
    {
      "auxiliary_loss_clip": 0.01334539,
      "auxiliary_loss_mlp": 0.01064122,
      "balance_loss_clip": 1.09280026,
      "balance_loss_mlp": 1.03727651,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 1.8379757802489076,
      "language_loss": 0.81080526,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83479184,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.626427412033081
    },
    {
      "auxiliary_loss_clip": 0.01329249,
      "auxiliary_loss_mlp": 0.01077297,
      "balance_loss_clip": 1.09112501,
      "balance_loss_mlp": 1.04975951,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 2.1774333955003575,
      "language_loss": 0.89272296,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91678834,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.7049341201782227
    },
    {
      "auxiliary_loss_clip": 0.01328911,
      "auxiliary_loss_mlp": 0.01069458,
      "balance_loss_clip": 1.09096456,
      "balance_loss_mlp": 1.04205179,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.998234233650083,
      "language_loss": 0.87629604,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90027976,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.6397783756256104
    },
    {
      "auxiliary_loss_clip": 0.01329721,
      "auxiliary_loss_mlp": 0.0106861,
      "balance_loss_clip": 1.09110284,
      "balance_loss_mlp": 1.04305148,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 1.9987228413248768,
      "language_loss": 0.874475,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.89845836,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.687316417694092
    },
    {
      "auxiliary_loss_clip": 0.01328492,
      "auxiliary_loss_mlp": 0.01072294,
      "balance_loss_clip": 1.08799934,
      "balance_loss_mlp": 1.0476172,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.366750868317763,
      "language_loss": 0.91559958,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.9396075,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.6888773441314697
    },
    {
      "auxiliary_loss_clip": 0.01333623,
      "auxiliary_loss_mlp": 0.01077333,
      "balance_loss_clip": 1.0954622,
      "balance_loss_mlp": 1.05154777,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 4.780878527150375,
      "language_loss": 0.92974204,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.95385158,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.6751065254211426
    },
    {
      "auxiliary_loss_clip": 0.01333766,
      "auxiliary_loss_mlp": 0.01070094,
      "balance_loss_clip": 1.09379029,
      "balance_loss_mlp": 1.04401064,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.6961802635285896,
      "language_loss": 0.8467474,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.87078595,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.668060302734375
    },
    {
      "auxiliary_loss_clip": 0.0132449,
      "auxiliary_loss_mlp": 0.01066103,
      "balance_loss_clip": 1.08677423,
      "balance_loss_mlp": 1.0404253,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 1.9701326047134609,
      "language_loss": 0.89644766,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92035365,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.6138956546783447
    },
    {
      "auxiliary_loss_clip": 0.013266,
      "auxiliary_loss_mlp": 0.01071995,
      "balance_loss_clip": 1.08998895,
      "balance_loss_mlp": 1.044958,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 2.3461174988018283,
      "language_loss": 0.84366316,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.86764914,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.699169635772705
    },
    {
      "auxiliary_loss_clip": 0.0133287,
      "auxiliary_loss_mlp": 0.01067408,
      "balance_loss_clip": 1.0915606,
      "balance_loss_mlp": 1.04021668,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 1.7245872075787525,
      "language_loss": 0.80360138,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.82760417,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.682603120803833
    },
    {
      "auxiliary_loss_clip": 0.01328474,
      "auxiliary_loss_mlp": 0.01065203,
      "balance_loss_clip": 1.09132111,
      "balance_loss_mlp": 1.03890514,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.0934208962977254,
      "language_loss": 0.82747018,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.85140705,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.7231874465942383
    },
    {
      "auxiliary_loss_clip": 0.01327485,
      "auxiliary_loss_mlp": 0.01066223,
      "balance_loss_clip": 1.08901417,
      "balance_loss_mlp": 1.03966284,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 2.267337908989615,
      "language_loss": 0.74237549,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.7663126,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 2.6859195232391357
    },
    {
      "auxiliary_loss_clip": 0.01328932,
      "auxiliary_loss_mlp": 0.01073904,
      "balance_loss_clip": 1.08695233,
      "balance_loss_mlp": 1.04730844,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 2.5183232028917635,
      "language_loss": 0.856812,
      "learning_rate": 4e-06,
      "loss": 0.8808403,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 3.59822416305542
    },
    {
      "auxiliary_loss_clip": 0.01327956,
      "auxiliary_loss_mlp": 0.01085146,
      "balance_loss_clip": 1.09198809,
      "balance_loss_mlp": 1.05917048,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 3.3249231532848507,
      "language_loss": 0.88415676,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.90828776,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 3.5388479232788086
    },
    {
      "auxiliary_loss_clip": 0.01320408,
      "auxiliary_loss_mlp": 0.01068542,
      "balance_loss_clip": 1.08436441,
      "balance_loss_mlp": 1.04264998,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 1.6178770783383485,
      "language_loss": 0.89081287,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91470242,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 4.531958341598511
    },
    {
      "auxiliary_loss_clip": 0.01320788,
      "auxiliary_loss_mlp": 0.01058686,
      "balance_loss_clip": 1.08705235,
      "balance_loss_mlp": 1.03321052,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 2.0124070493505717,
      "language_loss": 0.84988511,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.87367988,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.6529738903045654
    },
    {
      "auxiliary_loss_clip": 0.01328875,
      "auxiliary_loss_mlp": 0.0107228,
      "balance_loss_clip": 1.09229803,
      "balance_loss_mlp": 1.0452081,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 3.360548576622451,
      "language_loss": 0.96246898,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98648047,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 2.6530933380126953
    },
    {
      "auxiliary_loss_clip": 0.01323956,
      "auxiliary_loss_mlp": 0.01061945,
      "balance_loss_clip": 1.08559299,
      "balance_loss_mlp": 1.03662503,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 1.9224367540998473,
      "language_loss": 0.87627351,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.90013254,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.6173388957977295
    },
    {
      "auxiliary_loss_clip": 0.01325613,
      "auxiliary_loss_mlp": 0.01059839,
      "balance_loss_clip": 1.08782148,
      "balance_loss_mlp": 1.03360105,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.8725600542972165,
      "language_loss": 0.82231289,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84616745,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.6085612773895264
    },
    {
      "auxiliary_loss_clip": 0.01321779,
      "auxiliary_loss_mlp": 0.01065932,
      "balance_loss_clip": 1.08574879,
      "balance_loss_mlp": 1.04011166,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 2.628418314360437,
      "language_loss": 0.85879534,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.88267243,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.65468168258667
    },
    {
      "auxiliary_loss_clip": 0.01319293,
      "auxiliary_loss_mlp": 0.01053634,
      "balance_loss_clip": 1.13350284,
      "balance_loss_mlp": 1.03274906,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0055763096718673,
      "language_loss": 0.53476125,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55849057,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.2629518508911133
    },
    {
      "auxiliary_loss_clip": 0.01325855,
      "auxiliary_loss_mlp": 0.01058969,
      "balance_loss_clip": 1.09059811,
      "balance_loss_mlp": 1.03317249,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 3.4230621901132854,
      "language_loss": 0.86465079,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.88849902,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.634861946105957
    },
    {
      "auxiliary_loss_clip": 0.01321068,
      "auxiliary_loss_mlp": 0.01071819,
      "balance_loss_clip": 1.08917606,
      "balance_loss_mlp": 1.04647493,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 3.62420427810412,
      "language_loss": 0.8206659,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84459484,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 2.6655774116516113
    },
    {
      "auxiliary_loss_clip": 0.01326657,
      "auxiliary_loss_mlp": 0.01063737,
      "balance_loss_clip": 1.09090638,
      "balance_loss_mlp": 1.0376544,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 4.12364439235462,
      "language_loss": 0.97984093,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00374496,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 2.6675117015838623
    },
    {
      "auxiliary_loss_clip": 0.01324917,
      "auxiliary_loss_mlp": 0.01062388,
      "balance_loss_clip": 1.08807456,
      "balance_loss_mlp": 1.03578067,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.5314128276034293,
      "language_loss": 0.85968995,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88356292,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.632423162460327
    },
    {
      "auxiliary_loss_clip": 0.01323226,
      "auxiliary_loss_mlp": 0.0107028,
      "balance_loss_clip": 1.08640397,
      "balance_loss_mlp": 1.0439353,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 3.803522378064602,
      "language_loss": 0.80274165,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82667667,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.75121808052063
    },
    {
      "auxiliary_loss_clip": 0.01321763,
      "auxiliary_loss_mlp": 0.01063304,
      "balance_loss_clip": 1.08674335,
      "balance_loss_mlp": 1.03714907,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 2.2593566658610382,
      "language_loss": 0.89489818,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.91874886,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.6790049076080322
    },
    {
      "auxiliary_loss_clip": 0.01322157,
      "auxiliary_loss_mlp": 0.01072226,
      "balance_loss_clip": 1.08809042,
      "balance_loss_mlp": 1.04745436,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 2.0244784238178846,
      "language_loss": 0.80064666,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82459044,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.687027931213379
    },
    {
      "auxiliary_loss_clip": 0.0132329,
      "auxiliary_loss_mlp": 0.01061691,
      "balance_loss_clip": 1.08855629,
      "balance_loss_mlp": 1.03519034,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 2.776021398802981,
      "language_loss": 0.86790282,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.8917526,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.6232011318206787
    },
    {
      "auxiliary_loss_clip": 0.0132652,
      "auxiliary_loss_mlp": 0.01068223,
      "balance_loss_clip": 1.088148,
      "balance_loss_mlp": 1.04242659,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 3.2675831698480926,
      "language_loss": 0.78853977,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81248724,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.602123737335205
    },
    {
      "auxiliary_loss_clip": 0.01322681,
      "auxiliary_loss_mlp": 0.0106865,
      "balance_loss_clip": 1.08957887,
      "balance_loss_mlp": 1.04430747,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.4973045256274453,
      "language_loss": 0.79636657,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.82027984,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.7178800106048584
    },
    {
      "auxiliary_loss_clip": 0.01324119,
      "auxiliary_loss_mlp": 0.0106684,
      "balance_loss_clip": 1.08816087,
      "balance_loss_mlp": 1.04098356,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.573500216360628,
      "language_loss": 0.83933139,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.86324096,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.5950868129730225
    },
    {
      "auxiliary_loss_clip": 0.01328265,
      "auxiliary_loss_mlp": 0.01066551,
      "balance_loss_clip": 1.0930202,
      "balance_loss_mlp": 1.0407182,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 3.0219629024394172,
      "language_loss": 0.94734442,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.97129256,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.6705379486083984
    },
    {
      "auxiliary_loss_clip": 0.01325169,
      "auxiliary_loss_mlp": 0.01078926,
      "balance_loss_clip": 1.09030771,
      "balance_loss_mlp": 1.05280685,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 2.2940659382051614,
      "language_loss": 0.898067,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92210799,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.7130696773529053
    },
    {
      "auxiliary_loss_clip": 0.01323248,
      "auxiliary_loss_mlp": 0.01072959,
      "balance_loss_clip": 1.08999038,
      "balance_loss_mlp": 1.04790163,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 3.5473858806194625,
      "language_loss": 0.86687464,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89083672,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.7470593452453613
    },
    {
      "auxiliary_loss_clip": 0.01323858,
      "auxiliary_loss_mlp": 0.01067737,
      "balance_loss_clip": 1.08761597,
      "balance_loss_mlp": 1.04155815,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 2.2778765551524733,
      "language_loss": 0.63006496,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.65398091,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.864765167236328
    },
    {
      "auxiliary_loss_clip": 0.01322838,
      "auxiliary_loss_mlp": 0.01066048,
      "balance_loss_clip": 1.09083712,
      "balance_loss_mlp": 1.0402869,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 2.6568775006744265,
      "language_loss": 0.75592947,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.7798183,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.6612348556518555
    },
    {
      "auxiliary_loss_clip": 0.01326691,
      "auxiliary_loss_mlp": 0.01073133,
      "balance_loss_clip": 1.09275579,
      "balance_loss_mlp": 1.04747927,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.1686456987936054,
      "language_loss": 0.89623338,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.92023158,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.6370582580566406
    },
    {
      "auxiliary_loss_clip": 0.01321563,
      "auxiliary_loss_mlp": 0.01076941,
      "balance_loss_clip": 1.08941877,
      "balance_loss_mlp": 1.05100095,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.9423238749033893,
      "language_loss": 0.92874289,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.95272791,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 3.5640103816986084
    },
    {
      "auxiliary_loss_clip": 0.01322108,
      "auxiliary_loss_mlp": 0.0106039,
      "balance_loss_clip": 1.08742452,
      "balance_loss_mlp": 1.03529668,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 2.2372619885339993,
      "language_loss": 1.00153041,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02535534,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 3.610960006713867
    },
    {
      "auxiliary_loss_clip": 0.01321291,
      "auxiliary_loss_mlp": 0.01065971,
      "balance_loss_clip": 1.08609629,
      "balance_loss_mlp": 1.0411756,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 2.2697214977276796,
      "language_loss": 0.78663361,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81050622,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 4.497699975967407
    },
    {
      "auxiliary_loss_clip": 0.01321112,
      "auxiliary_loss_mlp": 0.01072575,
      "balance_loss_clip": 1.08727157,
      "balance_loss_mlp": 1.04617012,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 2.7450473194284997,
      "language_loss": 0.86590409,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.88984096,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 2.658735513687134
    },
    {
      "auxiliary_loss_clip": 0.01325957,
      "auxiliary_loss_mlp": 0.01065395,
      "balance_loss_clip": 1.09201121,
      "balance_loss_mlp": 1.03862023,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 1.8807310759220126,
      "language_loss": 0.79655337,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.820467,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 2.681607246398926
    },
    {
      "auxiliary_loss_clip": 0.01323512,
      "auxiliary_loss_mlp": 0.01066826,
      "balance_loss_clip": 1.09029043,
      "balance_loss_mlp": 1.04120779,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 1.9575523994810513,
      "language_loss": 0.78630686,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.81021017,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.7946791648864746
    },
    {
      "auxiliary_loss_clip": 0.01322846,
      "auxiliary_loss_mlp": 0.01065717,
      "balance_loss_clip": 1.08925176,
      "balance_loss_mlp": 1.04170823,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 2.7123377284950174,
      "language_loss": 0.81873286,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84261853,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.6687614917755127
    },
    {
      "auxiliary_loss_clip": 0.01318447,
      "auxiliary_loss_mlp": 0.01067055,
      "balance_loss_clip": 1.08707786,
      "balance_loss_mlp": 1.04126966,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 1.8106437359722554,
      "language_loss": 0.92072856,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94458359,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.6949539184570312
    },
    {
      "auxiliary_loss_clip": 0.01317917,
      "auxiliary_loss_mlp": 0.01068214,
      "balance_loss_clip": 1.08770132,
      "balance_loss_mlp": 1.04388344,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 2.283579054566723,
      "language_loss": 0.73864865,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.76250994,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.682826280593872
    },
    {
      "auxiliary_loss_clip": 0.01322845,
      "auxiliary_loss_mlp": 0.01063822,
      "balance_loss_clip": 1.08802819,
      "balance_loss_mlp": 1.03806138,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.092775959478182,
      "language_loss": 0.90731007,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93117678,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.6461944580078125
    },
    {
      "auxiliary_loss_clip": 0.0131787,
      "auxiliary_loss_mlp": 0.01062494,
      "balance_loss_clip": 1.08703804,
      "balance_loss_mlp": 1.0366019,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 1.69414458870202,
      "language_loss": 0.7377637,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.76156735,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.7384939193725586
    },
    {
      "auxiliary_loss_clip": 0.01320054,
      "auxiliary_loss_mlp": 0.01082423,
      "balance_loss_clip": 1.08971179,
      "balance_loss_mlp": 1.05779433,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 2.110435034955653,
      "language_loss": 0.75737274,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78139752,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 2.7713699340820312
    },
    {
      "auxiliary_loss_clip": 0.01323549,
      "auxiliary_loss_mlp": 0.01074058,
      "balance_loss_clip": 1.09286809,
      "balance_loss_mlp": 1.04941702,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.098926876668241,
      "language_loss": 0.96114737,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.9851234,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 2.686983585357666
    },
    {
      "auxiliary_loss_clip": 0.013257,
      "auxiliary_loss_mlp": 0.0106841,
      "balance_loss_clip": 1.08967614,
      "balance_loss_mlp": 1.04287553,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 2.0014051995798665,
      "language_loss": 0.82634783,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.85028899,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.8016090393066406
    },
    {
      "auxiliary_loss_clip": 0.01318493,
      "auxiliary_loss_mlp": 0.0106851,
      "balance_loss_clip": 1.08936989,
      "balance_loss_mlp": 1.04341626,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 1.836690275755057,
      "language_loss": 0.81235689,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83622682,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.745262861251831
    },
    {
      "auxiliary_loss_clip": 0.01320533,
      "auxiliary_loss_mlp": 0.01079291,
      "balance_loss_clip": 1.08920908,
      "balance_loss_mlp": 1.05313623,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 2.8624963773938683,
      "language_loss": 0.90525293,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.92925119,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.698057174682617
    },
    {
      "auxiliary_loss_clip": 0.01289716,
      "auxiliary_loss_mlp": 0.01026554,
      "balance_loss_clip": 1.12116969,
      "balance_loss_mlp": 1.00719452,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9876604892970776,
      "language_loss": 0.61166406,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63482678,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.2579591274261475
    },
    {
      "auxiliary_loss_clip": 0.01317193,
      "auxiliary_loss_mlp": 0.01068142,
      "balance_loss_clip": 1.08704138,
      "balance_loss_mlp": 1.04261959,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 2.273920476369607,
      "language_loss": 0.91847277,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94232613,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.6865241527557373
    },
    {
      "auxiliary_loss_clip": 0.01317713,
      "auxiliary_loss_mlp": 0.01078235,
      "balance_loss_clip": 1.08919632,
      "balance_loss_mlp": 1.05336797,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 2.648958977278746,
      "language_loss": 0.74703169,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77099109,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.8165476322174072
    },
    {
      "auxiliary_loss_clip": 0.01315793,
      "auxiliary_loss_mlp": 0.01072081,
      "balance_loss_clip": 1.08586311,
      "balance_loss_mlp": 1.0475719,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.2084037499142797,
      "language_loss": 0.85742784,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88130659,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.7321693897247314
    },
    {
      "auxiliary_loss_clip": 0.01279599,
      "auxiliary_loss_mlp": 0.01019617,
      "balance_loss_clip": 1.1130271,
      "balance_loss_mlp": 1.00025737,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 1.1030142824770583,
      "language_loss": 0.69442737,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71741951,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.3360049724578857
    },
    {
      "auxiliary_loss_clip": 0.0131538,
      "auxiliary_loss_mlp": 0.01056987,
      "balance_loss_clip": 1.08562028,
      "balance_loss_mlp": 1.03201282,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 2.617394759645052,
      "language_loss": 0.85467654,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.87840021,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.681971549987793
    },
    {
      "auxiliary_loss_clip": 0.01319749,
      "auxiliary_loss_mlp": 0.01064485,
      "balance_loss_clip": 1.08990335,
      "balance_loss_mlp": 1.03996372,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.304021239717147,
      "language_loss": 0.6586861,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.68252844,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.9116694927215576
    },
    {
      "auxiliary_loss_clip": 0.01316481,
      "auxiliary_loss_mlp": 0.01074857,
      "balance_loss_clip": 1.08947873,
      "balance_loss_mlp": 1.04938161,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 5.645985397557338,
      "language_loss": 0.9037497,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92766303,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.7138283252716064
    },
    {
      "auxiliary_loss_clip": 0.01269333,
      "auxiliary_loss_mlp": 0.01018247,
      "balance_loss_clip": 1.1055131,
      "balance_loss_mlp": 0.99898285,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.9200019911346663,
      "language_loss": 0.52294785,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54582363,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 3.045332670211792
    },
    {
      "auxiliary_loss_clip": 0.01313963,
      "auxiliary_loss_mlp": 0.01062728,
      "balance_loss_clip": 1.08505249,
      "balance_loss_mlp": 1.03782499,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 1.690762597575365,
      "language_loss": 0.86303419,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.88680112,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.7071921825408936
    },
    {
      "auxiliary_loss_clip": 0.01318291,
      "auxiliary_loss_mlp": 0.01062311,
      "balance_loss_clip": 1.08719039,
      "balance_loss_mlp": 1.03663313,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 3.8401469484759567,
      "language_loss": 0.89261353,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.91641951,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.60490345954895
    },
    {
      "auxiliary_loss_clip": 0.01314948,
      "auxiliary_loss_mlp": 0.01052983,
      "balance_loss_clip": 1.08488846,
      "balance_loss_mlp": 1.02760291,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 1.8723594888469313,
      "language_loss": 0.84716046,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87083983,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 4.511361360549927
    },
    {
      "auxiliary_loss_clip": 0.01321653,
      "auxiliary_loss_mlp": 0.0107181,
      "balance_loss_clip": 1.08944273,
      "balance_loss_mlp": 1.04647779,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 2.099514017576814,
      "language_loss": 0.81038976,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83432436,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 3.592745542526245
    },
    {
      "auxiliary_loss_clip": 0.01320803,
      "auxiliary_loss_mlp": 0.01074351,
      "balance_loss_clip": 1.08924747,
      "balance_loss_mlp": 1.05012786,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 1.832786259192636,
      "language_loss": 0.81570023,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83965176,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 4.3295698165893555
    },
    {
      "auxiliary_loss_clip": 0.01316519,
      "auxiliary_loss_mlp": 0.01077037,
      "balance_loss_clip": 1.08768296,
      "balance_loss_mlp": 1.05165744,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 1.8546594779545893,
      "language_loss": 0.8124631,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.83639866,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 2.727748155593872
    },
    {
      "auxiliary_loss_clip": 0.0131782,
      "auxiliary_loss_mlp": 0.0105527,
      "balance_loss_clip": 1.08951974,
      "balance_loss_mlp": 1.03068948,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 3.285025056599215,
      "language_loss": 0.924052,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.94778287,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.7080063819885254
    },
    {
      "auxiliary_loss_clip": 0.01317007,
      "auxiliary_loss_mlp": 0.01069693,
      "balance_loss_clip": 1.08748102,
      "balance_loss_mlp": 1.04599416,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 3.4939997117454715,
      "language_loss": 0.97519827,
      "learning_rate": 3.999489705569744e-06,
      "loss": 0.99906528,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.667457103729248
    },
    {
      "auxiliary_loss_clip": 0.01311319,
      "auxiliary_loss_mlp": 0.01057121,
      "balance_loss_clip": 1.08476472,
      "balance_loss_mlp": 1.03279078,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 2.2850686682001125,
      "language_loss": 0.8643949,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88807935,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.8387529850006104
    },
    {
      "auxiliary_loss_clip": 0.01318379,
      "auxiliary_loss_mlp": 0.01066744,
      "balance_loss_clip": 1.08935535,
      "balance_loss_mlp": 1.04337943,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 1.8100430479127776,
      "language_loss": 0.76217937,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.78603065,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.8004956245422363
    },
    {
      "auxiliary_loss_clip": 0.01310883,
      "auxiliary_loss_mlp": 0.01068636,
      "balance_loss_clip": 1.08485603,
      "balance_loss_mlp": 1.04417467,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 2.10814216679222,
      "language_loss": 0.81444204,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83823717,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.7258293628692627
    },
    {
      "auxiliary_loss_clip": 0.01313362,
      "auxiliary_loss_mlp": 0.010591,
      "balance_loss_clip": 1.08579135,
      "balance_loss_mlp": 1.03498399,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.3749996338283808,
      "language_loss": 0.83500063,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85872525,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.763798236846924
    },
    {
      "auxiliary_loss_clip": 0.01312154,
      "auxiliary_loss_mlp": 0.01066258,
      "balance_loss_clip": 1.08331394,
      "balance_loss_mlp": 1.04176021,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 3.5120446996678774,
      "language_loss": 0.85992211,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.88370627,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.753394842147827
    },
    {
      "auxiliary_loss_clip": 0.01318802,
      "auxiliary_loss_mlp": 0.01076152,
      "balance_loss_clip": 1.08790016,
      "balance_loss_mlp": 1.05149984,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 3.573574028365135,
      "language_loss": 0.82751089,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.85146046,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 2.7045633792877197
    },
    {
      "auxiliary_loss_clip": 0.01313333,
      "auxiliary_loss_mlp": 0.01057185,
      "balance_loss_clip": 1.08662617,
      "balance_loss_mlp": 1.03297329,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 3.0196864729065753,
      "language_loss": 0.83115232,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85485744,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 2.6907410621643066
    },
    {
      "auxiliary_loss_clip": 0.01312125,
      "auxiliary_loss_mlp": 0.01071842,
      "balance_loss_clip": 1.08623862,
      "balance_loss_mlp": 1.04815555,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 1.9886595267986127,
      "language_loss": 0.79425728,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81809694,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.7331595420837402
    },
    {
      "auxiliary_loss_clip": 0.013154,
      "auxiliary_loss_mlp": 0.01064394,
      "balance_loss_clip": 1.08678091,
      "balance_loss_mlp": 1.04080248,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 1.929357726772405,
      "language_loss": 0.83790457,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.86170244,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.7512638568878174
    },
    {
      "auxiliary_loss_clip": 0.01311875,
      "auxiliary_loss_mlp": 0.01060508,
      "balance_loss_clip": 1.08382154,
      "balance_loss_mlp": 1.03673768,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 2.601853502626988,
      "language_loss": 0.83151728,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85524112,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.7298166751861572
    },
    {
      "auxiliary_loss_clip": 0.0131271,
      "auxiliary_loss_mlp": 0.01060011,
      "balance_loss_clip": 1.08576334,
      "balance_loss_mlp": 1.03639603,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 2.0867710402411666,
      "language_loss": 0.77480453,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.79853177,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.6823902130126953
    },
    {
      "auxiliary_loss_clip": 0.012592,
      "auxiliary_loss_mlp": 0.01023644,
      "balance_loss_clip": 1.10264194,
      "balance_loss_mlp": 1.00542915,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9528655352893038,
      "language_loss": 0.62399149,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64681995,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.255526304244995
    },
    {
      "auxiliary_loss_clip": 0.01256797,
      "auxiliary_loss_mlp": 0.01020968,
      "balance_loss_clip": 1.10106575,
      "balance_loss_mlp": 1.00284851,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.216440385511086,
      "language_loss": 0.67060792,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.6933856,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.1571543216705322
    },
    {
      "auxiliary_loss_clip": 0.01314661,
      "auxiliary_loss_mlp": 0.01073088,
      "balance_loss_clip": 1.09028387,
      "balance_loss_mlp": 1.04955673,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 2.1617484881968108,
      "language_loss": 0.82003027,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.84390777,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.7309353351593018
    },
    {
      "auxiliary_loss_clip": 0.01313408,
      "auxiliary_loss_mlp": 0.01063148,
      "balance_loss_clip": 1.08763611,
      "balance_loss_mlp": 1.04011703,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 1.597343574166061,
      "language_loss": 0.91040683,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.93417239,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.7732057571411133
    },
    {
      "auxiliary_loss_clip": 0.01251088,
      "auxiliary_loss_mlp": 0.0102007,
      "balance_loss_clip": 1.09694231,
      "balance_loss_mlp": 1.00195062,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.9080004712426474,
      "language_loss": 0.59802544,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.62073708,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.168935775756836
    },
    {
      "auxiliary_loss_clip": 0.01315532,
      "auxiliary_loss_mlp": 0.0106687,
      "balance_loss_clip": 1.08790207,
      "balance_loss_mlp": 1.04137087,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 2.2646251698061035,
      "language_loss": 0.83040237,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85422641,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.8366591930389404
    },
    {
      "auxiliary_loss_clip": 0.01308052,
      "auxiliary_loss_mlp": 0.01049143,
      "balance_loss_clip": 1.08210874,
      "balance_loss_mlp": 1.02615976,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.510934566558406,
      "language_loss": 0.89699757,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.92056954,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.7013943195343018
    },
    {
      "auxiliary_loss_clip": 0.01306366,
      "auxiliary_loss_mlp": 0.01059775,
      "balance_loss_clip": 1.08141208,
      "balance_loss_mlp": 1.03635025,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 2.393072414944594,
      "language_loss": 0.74950385,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.77316523,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.729597330093384
    },
    {
      "auxiliary_loss_clip": 0.01313726,
      "auxiliary_loss_mlp": 0.01080604,
      "balance_loss_clip": 1.08782375,
      "balance_loss_mlp": 1.05691767,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 2.3167488415121777,
      "language_loss": 0.8831619,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90710521,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 3.821798324584961
    },
    {
      "auxiliary_loss_clip": 0.013134,
      "auxiliary_loss_mlp": 0.01060082,
      "balance_loss_clip": 1.08665872,
      "balance_loss_mlp": 1.03746796,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 2.476713900224869,
      "language_loss": 0.90700901,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.93074387,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 2.8059892654418945
    },
    {
      "auxiliary_loss_clip": 0.0131155,
      "auxiliary_loss_mlp": 0.01063319,
      "balance_loss_clip": 1.08712769,
      "balance_loss_mlp": 1.04072928,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.283655493714867,
      "language_loss": 0.75729191,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.78104061,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 3.635004758834839
    },
    {
      "auxiliary_loss_clip": 0.01308161,
      "auxiliary_loss_mlp": 0.01068551,
      "balance_loss_clip": 1.08619261,
      "balance_loss_mlp": 1.0446496,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 2.0577645270276674,
      "language_loss": 0.86685491,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.89062196,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 4.98781943321228
    },
    {
      "auxiliary_loss_clip": 0.01309198,
      "auxiliary_loss_mlp": 0.01062785,
      "balance_loss_clip": 1.08673406,
      "balance_loss_mlp": 1.04000378,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.2891423543437615,
      "language_loss": 0.82645071,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.85017049,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.743177652359009
    },
    {
      "auxiliary_loss_clip": 0.01310562,
      "auxiliary_loss_mlp": 0.01064225,
      "balance_loss_clip": 1.08475745,
      "balance_loss_mlp": 1.04097879,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 3.082906193171117,
      "language_loss": 0.75603271,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.77978063,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 2.7834603786468506
    },
    {
      "auxiliary_loss_clip": 0.01307984,
      "auxiliary_loss_mlp": 0.0106369,
      "balance_loss_clip": 1.08405399,
      "balance_loss_mlp": 1.04063535,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.425682165500368,
      "language_loss": 0.84480548,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.86852223,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.7606918811798096
    },
    {
      "auxiliary_loss_clip": 0.01307143,
      "auxiliary_loss_mlp": 0.01066614,
      "balance_loss_clip": 1.08479512,
      "balance_loss_mlp": 1.04320109,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 2.7196764034101575,
      "language_loss": 0.89370859,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.9174462,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.7913477420806885
    },
    {
      "auxiliary_loss_clip": 0.01306352,
      "auxiliary_loss_mlp": 0.01059459,
      "balance_loss_clip": 1.08273911,
      "balance_loss_mlp": 1.0369519,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.7071019171983308,
      "language_loss": 0.87105298,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.89471114,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.7657151222229004
    },
    {
      "auxiliary_loss_clip": 0.01305734,
      "auxiliary_loss_mlp": 0.01060239,
      "balance_loss_clip": 1.08385599,
      "balance_loss_mlp": 1.03706503,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 4.023973498850077,
      "language_loss": 0.85500574,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87866557,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.8008317947387695
    },
    {
      "auxiliary_loss_clip": 0.013026,
      "auxiliary_loss_mlp": 0.01056539,
      "balance_loss_clip": 1.08045435,
      "balance_loss_mlp": 1.03396058,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 1.8636849512976839,
      "language_loss": 0.82970726,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.8532986,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.6948704719543457
    },
    {
      "auxiliary_loss_clip": 0.01306297,
      "auxiliary_loss_mlp": 0.01067248,
      "balance_loss_clip": 1.08348393,
      "balance_loss_mlp": 1.04501569,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 3.2954249826319177,
      "language_loss": 0.73085696,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75459242,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.884826421737671
    },
    {
      "auxiliary_loss_clip": 0.01303867,
      "auxiliary_loss_mlp": 0.01055257,
      "balance_loss_clip": 1.08044529,
      "balance_loss_mlp": 1.03302407,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.428915147298781,
      "language_loss": 0.83820903,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86180025,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.759138822555542
    },
    {
      "auxiliary_loss_clip": 0.01305375,
      "auxiliary_loss_mlp": 0.01059119,
      "balance_loss_clip": 1.08261299,
      "balance_loss_mlp": 1.03608811,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 2.465196035488566,
      "language_loss": 0.87754071,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.90118563,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 2.759354591369629
    },
    {
      "auxiliary_loss_clip": 0.01311315,
      "auxiliary_loss_mlp": 0.0106279,
      "balance_loss_clip": 1.08841968,
      "balance_loss_mlp": 1.04120147,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.696388828987813,
      "language_loss": 0.73411208,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.75785309,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 2.6773743629455566
    },
    {
      "auxiliary_loss_clip": 0.01307663,
      "auxiliary_loss_mlp": 0.01062918,
      "balance_loss_clip": 1.08725023,
      "balance_loss_mlp": 1.04049456,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 2.6962587892967744,
      "language_loss": 0.90722132,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.93092716,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.7878575325012207
    },
    {
      "auxiliary_loss_clip": 0.01304776,
      "auxiliary_loss_mlp": 0.01054735,
      "balance_loss_clip": 1.08235705,
      "balance_loss_mlp": 1.03226376,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.2108260745977,
      "language_loss": 0.79997599,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.82357109,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.7431252002716064
    },
    {
      "auxiliary_loss_clip": 0.01304465,
      "auxiliary_loss_mlp": 0.01066004,
      "balance_loss_clip": 1.08306181,
      "balance_loss_mlp": 1.04360414,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 2.11061994895343,
      "language_loss": 0.77810282,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.80180752,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.776367425918579
    },
    {
      "auxiliary_loss_clip": 0.01305195,
      "auxiliary_loss_mlp": 0.01063389,
      "balance_loss_clip": 1.08559155,
      "balance_loss_mlp": 1.04182386,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 4.444775144664832,
      "language_loss": 0.83528215,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85896802,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.7372703552246094
    },
    {
      "auxiliary_loss_clip": 0.01306252,
      "auxiliary_loss_mlp": 0.01068526,
      "balance_loss_clip": 1.08248949,
      "balance_loss_mlp": 1.04555464,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 2.770425351241236,
      "language_loss": 0.91613495,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.93988281,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.7441864013671875
    },
    {
      "auxiliary_loss_clip": 0.0130458,
      "auxiliary_loss_mlp": 0.0105852,
      "balance_loss_clip": 1.08195376,
      "balance_loss_mlp": 1.03467858,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 2.1827232787304074,
      "language_loss": 0.84958255,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.87321353,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.810096025466919
    },
    {
      "auxiliary_loss_clip": 0.01306118,
      "auxiliary_loss_mlp": 0.01061803,
      "balance_loss_clip": 1.08513725,
      "balance_loss_mlp": 1.03868878,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 1.8401850111862401,
      "language_loss": 0.87805331,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90173256,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.879423141479492
    },
    {
      "auxiliary_loss_clip": 0.01307398,
      "auxiliary_loss_mlp": 0.01069676,
      "balance_loss_clip": 1.08533716,
      "balance_loss_mlp": 1.04803944,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 2.224110517314045,
      "language_loss": 0.94985616,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.97362691,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.8449652194976807
    },
    {
      "auxiliary_loss_clip": 0.01308544,
      "auxiliary_loss_mlp": 0.01080105,
      "balance_loss_clip": 1.0847466,
      "balance_loss_mlp": 1.05700254,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.25042103146398,
      "language_loss": 0.79032624,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81421268,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.776538133621216
    },
    {
      "auxiliary_loss_clip": 0.01300707,
      "auxiliary_loss_mlp": 0.0105857,
      "balance_loss_clip": 1.07964826,
      "balance_loss_mlp": 1.03606355,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.1618391462617033,
      "language_loss": 0.84322441,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.86681718,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.887808322906494
    },
    {
      "auxiliary_loss_clip": 0.01302531,
      "auxiliary_loss_mlp": 0.01061996,
      "balance_loss_clip": 1.08265471,
      "balance_loss_mlp": 1.03989482,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 1.6080898760575693,
      "language_loss": 0.89167172,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.91531694,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.9426677227020264
    },
    {
      "auxiliary_loss_clip": 0.01304319,
      "auxiliary_loss_mlp": 0.010628,
      "balance_loss_clip": 1.08155835,
      "balance_loss_mlp": 1.04032958,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 2.675552227400016,
      "language_loss": 0.90040159,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.92407274,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.9474055767059326
    },
    {
      "auxiliary_loss_clip": 0.01299997,
      "auxiliary_loss_mlp": 0.01065532,
      "balance_loss_clip": 1.079427,
      "balance_loss_mlp": 1.04377627,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 2.345512648114349,
      "language_loss": 0.81065631,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.8343116,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 3.661593198776245
    },
    {
      "auxiliary_loss_clip": 0.01299461,
      "auxiliary_loss_mlp": 0.01069705,
      "balance_loss_clip": 1.07839632,
      "balance_loss_mlp": 1.04825902,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.6475529192907945,
      "language_loss": 0.8125273,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83621895,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 4.186137676239014
    },
    {
      "auxiliary_loss_clip": 0.01307839,
      "auxiliary_loss_mlp": 0.01067602,
      "balance_loss_clip": 1.08619905,
      "balance_loss_mlp": 1.04523838,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 1.8757236823451373,
      "language_loss": 0.83347529,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85722971,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 3.7313246726989746
    },
    {
      "auxiliary_loss_clip": 0.01300766,
      "auxiliary_loss_mlp": 0.01058312,
      "balance_loss_clip": 1.08021045,
      "balance_loss_mlp": 1.03684258,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 1.8881609556047723,
      "language_loss": 0.81823587,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84182668,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 4.581888437271118
    },
    {
      "auxiliary_loss_clip": 0.0130601,
      "auxiliary_loss_mlp": 0.01062046,
      "balance_loss_clip": 1.08613074,
      "balance_loss_mlp": 1.03909874,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.4071107708431447,
      "language_loss": 0.84765697,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.87133753,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 2.8487353324890137
    },
    {
      "auxiliary_loss_clip": 0.01298683,
      "auxiliary_loss_mlp": 0.01065259,
      "balance_loss_clip": 1.08099985,
      "balance_loss_mlp": 1.04291952,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 1.9999089800113563,
      "language_loss": 0.86126244,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88490176,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.8029441833496094
    },
    {
      "auxiliary_loss_clip": 0.01301987,
      "auxiliary_loss_mlp": 0.01060233,
      "balance_loss_clip": 1.08278143,
      "balance_loss_mlp": 1.03791761,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 2.0380236139052696,
      "language_loss": 0.9599191,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98354125,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.7862765789031982
    },
    {
      "auxiliary_loss_clip": 0.01301399,
      "auxiliary_loss_mlp": 0.01061211,
      "balance_loss_clip": 1.08097005,
      "balance_loss_mlp": 1.03936028,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 2.5153155869872963,
      "language_loss": 0.76497507,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.78860116,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.739643096923828
    },
    {
      "auxiliary_loss_clip": 0.01298476,
      "auxiliary_loss_mlp": 0.01054598,
      "balance_loss_clip": 1.08033645,
      "balance_loss_mlp": 1.03351045,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.7054527306491407,
      "language_loss": 0.84745812,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.87098885,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.786482334136963
    },
    {
      "auxiliary_loss_clip": 0.0130037,
      "auxiliary_loss_mlp": 0.01066211,
      "balance_loss_clip": 1.08118105,
      "balance_loss_mlp": 1.04416966,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 2.015176159517788,
      "language_loss": 0.80895352,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83261931,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.746842622756958
    },
    {
      "auxiliary_loss_clip": 0.01300242,
      "auxiliary_loss_mlp": 0.0106447,
      "balance_loss_clip": 1.07918453,
      "balance_loss_mlp": 1.04271448,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 5.112202410901094,
      "language_loss": 0.82266772,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84631479,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.703413486480713
    },
    {
      "auxiliary_loss_clip": 0.0130488,
      "auxiliary_loss_mlp": 0.01048836,
      "balance_loss_clip": 1.082183,
      "balance_loss_mlp": 1.02681768,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.031613874769078,
      "language_loss": 0.83080167,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85433888,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.7634379863739014
    },
    {
      "auxiliary_loss_clip": 0.01305505,
      "auxiliary_loss_mlp": 0.01064038,
      "balance_loss_clip": 1.08512211,
      "balance_loss_mlp": 1.04184151,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.36196430208335,
      "language_loss": 0.81168079,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.83537614,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 2.7115297317504883
    },
    {
      "auxiliary_loss_clip": 0.0129999,
      "auxiliary_loss_mlp": 0.01057496,
      "balance_loss_clip": 1.08238316,
      "balance_loss_mlp": 1.03479886,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 1.7152420413408684,
      "language_loss": 0.88808757,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.91166246,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.7616913318634033
    },
    {
      "auxiliary_loss_clip": 0.0130295,
      "auxiliary_loss_mlp": 0.01055326,
      "balance_loss_clip": 1.08200765,
      "balance_loss_mlp": 1.03380847,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 2.896197691613399,
      "language_loss": 0.78897095,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.8125537,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.8414306640625
    },
    {
      "auxiliary_loss_clip": 0.0130594,
      "auxiliary_loss_mlp": 0.01069342,
      "balance_loss_clip": 1.08346272,
      "balance_loss_mlp": 1.0471096,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 2.429228854378308,
      "language_loss": 0.89165407,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91540694,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 2.7003798484802246
    },
    {
      "auxiliary_loss_clip": 0.0130422,
      "auxiliary_loss_mlp": 0.01064042,
      "balance_loss_clip": 1.08245528,
      "balance_loss_mlp": 1.04319239,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 1.7572909647736008,
      "language_loss": 0.80081582,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82449841,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.729684591293335
    },
    {
      "auxiliary_loss_clip": 0.012991,
      "auxiliary_loss_mlp": 0.01059072,
      "balance_loss_clip": 1.07902133,
      "balance_loss_mlp": 1.03747165,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 2.211203970772327,
      "language_loss": 0.73391181,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.7574935,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 2.8014276027679443
    },
    {
      "auxiliary_loss_clip": 0.01305514,
      "auxiliary_loss_mlp": 0.01072297,
      "balance_loss_clip": 1.08411372,
      "balance_loss_mlp": 1.04980254,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 2.2585814104012734,
      "language_loss": 0.92336059,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.94713867,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.713834524154663
    },
    {
      "auxiliary_loss_clip": 0.01239225,
      "auxiliary_loss_mlp": 0.0105411,
      "balance_loss_clip": 1.09220409,
      "balance_loss_mlp": 1.03808868,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 0.985644250171076,
      "language_loss": 0.60259712,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62553048,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.187532663345337
    },
    {
      "auxiliary_loss_clip": 0.01240436,
      "auxiliary_loss_mlp": 0.01036521,
      "balance_loss_clip": 1.09328532,
      "balance_loss_mlp": 1.0204041,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.9403081416162344,
      "language_loss": 0.57804239,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60081196,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.3417928218841553
    },
    {
      "auxiliary_loss_clip": 0.01303439,
      "auxiliary_loss_mlp": 0.01055072,
      "balance_loss_clip": 1.08280194,
      "balance_loss_mlp": 1.03291106,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 1.7229937043583072,
      "language_loss": 0.77298784,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.79657298,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.796447992324829
    },
    {
      "auxiliary_loss_clip": 0.01303372,
      "auxiliary_loss_mlp": 0.01057229,
      "balance_loss_clip": 1.08300948,
      "balance_loss_mlp": 1.03589034,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.495847909898285,
      "language_loss": 0.65897369,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.6825797,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.773810386657715
    },
    {
      "auxiliary_loss_clip": 0.0129958,
      "auxiliary_loss_mlp": 0.01055049,
      "balance_loss_clip": 1.08120584,
      "balance_loss_mlp": 1.03446162,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 1.877991625011866,
      "language_loss": 0.91214526,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.9356916,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.7045955657958984
    },
    {
      "auxiliary_loss_clip": 0.01298052,
      "auxiliary_loss_mlp": 0.01071118,
      "balance_loss_clip": 1.08247852,
      "balance_loss_mlp": 1.04999423,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 2.544744965089305,
      "language_loss": 0.86957765,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.8932693,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.7172420024871826
    },
    {
      "auxiliary_loss_clip": 0.01295579,
      "auxiliary_loss_mlp": 0.0106892,
      "balance_loss_clip": 1.08241677,
      "balance_loss_mlp": 1.04824901,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 2.1092563019367243,
      "language_loss": 0.77518237,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.79882729,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.7907798290252686
    },
    {
      "auxiliary_loss_clip": 0.01307972,
      "auxiliary_loss_mlp": 0.01067523,
      "balance_loss_clip": 1.08557749,
      "balance_loss_mlp": 1.04570746,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 3.2425605281854764,
      "language_loss": 0.74529064,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76904553,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 3.7200708389282227
    },
    {
      "auxiliary_loss_clip": 0.01231443,
      "auxiliary_loss_mlp": 0.010238,
      "balance_loss_clip": 1.08650243,
      "balance_loss_mlp": 1.00768244,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8247812469098866,
      "language_loss": 0.60053802,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.62309051,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 4.339107990264893
    },
    {
      "auxiliary_loss_clip": 0.01298676,
      "auxiliary_loss_mlp": 0.01060494,
      "balance_loss_clip": 1.08090019,
      "balance_loss_mlp": 1.0389533,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 1.8168025270283124,
      "language_loss": 0.75171858,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77531028,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 3.6038684844970703
    },
    {
      "auxiliary_loss_clip": 0.01299038,
      "auxiliary_loss_mlp": 0.01077142,
      "balance_loss_clip": 1.081375,
      "balance_loss_mlp": 1.05556512,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 1.9412743610068113,
      "language_loss": 0.82465464,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.84841645,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 4.02632737159729
    },
    {
      "auxiliary_loss_clip": 0.01303517,
      "auxiliary_loss_mlp": 0.01070097,
      "balance_loss_clip": 1.08547115,
      "balance_loss_mlp": 1.04859161,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 1.8357192773391977,
      "language_loss": 0.78038728,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.8041234,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 2.765868902206421
    },
    {
      "auxiliary_loss_clip": 0.01299083,
      "auxiliary_loss_mlp": 0.01047429,
      "balance_loss_clip": 1.08254457,
      "balance_loss_mlp": 1.02656746,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 1.8318318340916997,
      "language_loss": 0.82801414,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.85147923,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.7272448539733887
    },
    {
      "auxiliary_loss_clip": 0.01296671,
      "auxiliary_loss_mlp": 0.01065672,
      "balance_loss_clip": 1.07976961,
      "balance_loss_mlp": 1.04354692,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 2.3734939323255975,
      "language_loss": 0.9125222,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.9361456,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.6354751586914062
    },
    {
      "auxiliary_loss_clip": 0.0129952,
      "auxiliary_loss_mlp": 0.01071006,
      "balance_loss_clip": 1.08143485,
      "balance_loss_mlp": 1.04926193,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 2.3030333426905822,
      "language_loss": 0.81515145,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.8388567,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.740133047103882
    },
    {
      "auxiliary_loss_clip": 0.01301414,
      "auxiliary_loss_mlp": 0.01059946,
      "balance_loss_clip": 1.0825814,
      "balance_loss_mlp": 1.03746319,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 2.021095943779384,
      "language_loss": 0.82896596,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.85257959,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.691093921661377
    },
    {
      "auxiliary_loss_clip": 0.01294315,
      "auxiliary_loss_mlp": 0.01066455,
      "balance_loss_clip": 1.08001554,
      "balance_loss_mlp": 1.04559326,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.476989223785034,
      "language_loss": 0.80427915,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82788682,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.857544422149658
    },
    {
      "auxiliary_loss_clip": 0.01300321,
      "auxiliary_loss_mlp": 0.01059756,
      "balance_loss_clip": 1.08176577,
      "balance_loss_mlp": 1.03794038,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 2.6784242365838717,
      "language_loss": 0.89645201,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.92005271,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.6760802268981934
    },
    {
      "auxiliary_loss_clip": 0.01218102,
      "auxiliary_loss_mlp": 0.01029803,
      "balance_loss_clip": 1.07646561,
      "balance_loss_mlp": 1.01416254,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8671329654157214,
      "language_loss": 0.6387763,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66125536,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.323683261871338
    },
    {
      "auxiliary_loss_clip": 0.01304822,
      "auxiliary_loss_mlp": 0.01059008,
      "balance_loss_clip": 1.08773148,
      "balance_loss_mlp": 1.03714466,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 2.469634407072093,
      "language_loss": 0.81678945,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.84042776,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.7502825260162354
    },
    {
      "auxiliary_loss_clip": 0.01301762,
      "auxiliary_loss_mlp": 0.0105514,
      "balance_loss_clip": 1.08319759,
      "balance_loss_mlp": 1.03369403,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 1.9578657554630094,
      "language_loss": 0.87632799,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.89989698,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 2.6881191730499268
    },
    {
      "auxiliary_loss_clip": 0.0129387,
      "auxiliary_loss_mlp": 0.01060433,
      "balance_loss_clip": 1.07794344,
      "balance_loss_mlp": 1.03877246,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 1.8408660893592135,
      "language_loss": 0.92664367,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.95018673,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.805147886276245
    },
    {
      "auxiliary_loss_clip": 0.0130117,
      "auxiliary_loss_mlp": 0.01063718,
      "balance_loss_clip": 1.08685005,
      "balance_loss_mlp": 1.04210508,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 36.046821514440026,
      "language_loss": 0.7773487,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.80099761,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 2.774437189102173
    },
    {
      "auxiliary_loss_clip": 0.01300496,
      "auxiliary_loss_mlp": 0.01055231,
      "balance_loss_clip": 1.08557272,
      "balance_loss_mlp": 1.03277218,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 2.2435854030148845,
      "language_loss": 0.86772102,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89127827,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 2.7121527194976807
    },
    {
      "auxiliary_loss_clip": 0.01300928,
      "auxiliary_loss_mlp": 0.01062399,
      "balance_loss_clip": 1.08493543,
      "balance_loss_mlp": 1.04061913,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 3.036395430501904,
      "language_loss": 0.82860404,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.85223728,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.7156338691711426
    },
    {
      "auxiliary_loss_clip": 0.01294427,
      "auxiliary_loss_mlp": 0.01057869,
      "balance_loss_clip": 1.08134723,
      "balance_loss_mlp": 1.03598213,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 2.159826298269361,
      "language_loss": 0.8868925,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.91041541,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 2.7330141067504883
    },
    {
      "auxiliary_loss_clip": 0.01296627,
      "auxiliary_loss_mlp": 0.01054714,
      "balance_loss_clip": 1.08340573,
      "balance_loss_mlp": 1.03367305,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 2.5186016810366665,
      "language_loss": 0.81773323,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.84124666,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.8162965774536133
    },
    {
      "auxiliary_loss_clip": 0.01295155,
      "auxiliary_loss_mlp": 0.01062361,
      "balance_loss_clip": 1.07979548,
      "balance_loss_mlp": 1.04159427,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 3.326972525093555,
      "language_loss": 0.86486411,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88843924,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.8045902252197266
    },
    {
      "auxiliary_loss_clip": 0.01296729,
      "auxiliary_loss_mlp": 0.01067682,
      "balance_loss_clip": 1.08001196,
      "balance_loss_mlp": 1.04620028,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 2.418068404146797,
      "language_loss": 0.9416489,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.96529293,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.845588207244873
    },
    {
      "auxiliary_loss_clip": 0.0121522,
      "auxiliary_loss_mlp": 0.01021066,
      "balance_loss_clip": 1.07484472,
      "balance_loss_mlp": 1.00590253,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8943154822960663,
      "language_loss": 0.6314038,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65376663,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.3108067512512207
    },
    {
      "auxiliary_loss_clip": 0.01293793,
      "auxiliary_loss_mlp": 0.01062806,
      "balance_loss_clip": 1.08061755,
      "balance_loss_mlp": 1.04341066,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 2.1374682061747716,
      "language_loss": 0.79250073,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81606674,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.8217215538024902
    },
    {
      "auxiliary_loss_clip": 0.01292562,
      "auxiliary_loss_mlp": 0.01065845,
      "balance_loss_clip": 1.07982242,
      "balance_loss_mlp": 1.04530525,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 2.5430162463719994,
      "language_loss": 0.78568745,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.80927151,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.794588088989258
    },
    {
      "auxiliary_loss_clip": 0.01296698,
      "auxiliary_loss_mlp": 0.01061128,
      "balance_loss_clip": 1.07797432,
      "balance_loss_mlp": 1.03938472,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 2.1433920325949645,
      "language_loss": 0.61776197,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.64134026,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.8070385456085205
    },
    {
      "auxiliary_loss_clip": 0.01295057,
      "auxiliary_loss_mlp": 0.01063814,
      "balance_loss_clip": 1.08058631,
      "balance_loss_mlp": 1.04306006,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 2.1611466198288385,
      "language_loss": 0.73808461,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.76167333,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.7381701469421387
    },
    {
      "auxiliary_loss_clip": 0.01295502,
      "auxiliary_loss_mlp": 0.01062425,
      "balance_loss_clip": 1.08143938,
      "balance_loss_mlp": 1.04236197,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 1.6371908801135844,
      "language_loss": 0.86243308,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.88601232,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 5.016650676727295
    },
    {
      "auxiliary_loss_clip": 0.01299312,
      "auxiliary_loss_mlp": 0.01067403,
      "balance_loss_clip": 1.08303428,
      "balance_loss_mlp": 1.04539657,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 3.0103757969915748,
      "language_loss": 0.78717828,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81084538,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 3.6577329635620117
    },
    {
      "auxiliary_loss_clip": 0.01292888,
      "auxiliary_loss_mlp": 0.0105238,
      "balance_loss_clip": 1.07714343,
      "balance_loss_mlp": 1.03207827,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 2.147704450284593,
      "language_loss": 0.86735439,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.89080709,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 2.7041587829589844
    },
    {
      "auxiliary_loss_clip": 0.01294923,
      "auxiliary_loss_mlp": 0.01064632,
      "balance_loss_clip": 1.08086157,
      "balance_loss_mlp": 1.04346085,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 2.3163946704791774,
      "language_loss": 0.97729373,
      "learning_rate": 3.996069093204175e-06,
      "loss": 1.0008893,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 3.9674177169799805
    },
    {
      "auxiliary_loss_clip": 0.01300716,
      "auxiliary_loss_mlp": 0.01059936,
      "balance_loss_clip": 1.08479512,
      "balance_loss_mlp": 1.03858531,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 2.9152168926512707,
      "language_loss": 0.87917984,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.90278637,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 2.78570294380188
    },
    {
      "auxiliary_loss_clip": 0.01296021,
      "auxiliary_loss_mlp": 0.01055598,
      "balance_loss_clip": 1.08021772,
      "balance_loss_mlp": 1.03509426,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 2.2774229411581945,
      "language_loss": 0.82069218,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84420836,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 2.8887219429016113
    },
    {
      "auxiliary_loss_clip": 0.01296458,
      "auxiliary_loss_mlp": 0.01066028,
      "balance_loss_clip": 1.0806427,
      "balance_loss_mlp": 1.04477251,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 1.9582622893654724,
      "language_loss": 0.7977736,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.8213985,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 2.8335537910461426
    },
    {
      "auxiliary_loss_clip": 0.01288851,
      "auxiliary_loss_mlp": 0.01063873,
      "balance_loss_clip": 1.07860112,
      "balance_loss_mlp": 1.04340506,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.516679362228342,
      "language_loss": 0.80763978,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.83116704,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.728346824645996
    },
    {
      "auxiliary_loss_clip": 0.01213464,
      "auxiliary_loss_mlp": 0.01015647,
      "balance_loss_clip": 1.03963947,
      "balance_loss_mlp": 1.00067449,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9178013081078011,
      "language_loss": 0.5961535,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61844456,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.4919590950012207
    },
    {
      "auxiliary_loss_clip": 0.01292676,
      "auxiliary_loss_mlp": 0.01059285,
      "balance_loss_clip": 1.04415894,
      "balance_loss_mlp": 1.03901958,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 2.0113300432579355,
      "language_loss": 0.82144356,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.84496313,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.8074138164520264
    },
    {
      "auxiliary_loss_clip": 0.01294529,
      "auxiliary_loss_mlp": 0.01065327,
      "balance_loss_clip": 1.08147621,
      "balance_loss_mlp": 1.0438931,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 2.0955460711803515,
      "language_loss": 0.97296882,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.99656743,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.712995767593384
    },
    {
      "auxiliary_loss_clip": 0.01280112,
      "auxiliary_loss_mlp": 0.01063151,
      "balance_loss_clip": 1.00061345,
      "balance_loss_mlp": 1.04220557,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 2.5380615644702016,
      "language_loss": 0.85832769,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.8817603,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.8007326126098633
    },
    {
      "auxiliary_loss_clip": 0.012934,
      "auxiliary_loss_mlp": 0.01061208,
      "balance_loss_clip": 1.07975864,
      "balance_loss_mlp": 1.03882074,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.283471086397087,
      "language_loss": 0.90999115,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93353724,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.690749168395996
    },
    {
      "auxiliary_loss_clip": 0.01278343,
      "auxiliary_loss_mlp": 0.01062933,
      "balance_loss_clip": 0.99774724,
      "balance_loss_mlp": 1.04140389,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 2.468362324790034,
      "language_loss": 0.80492264,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.8283354,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 2.741971492767334
    },
    {
      "auxiliary_loss_clip": 0.0128074,
      "auxiliary_loss_mlp": 0.01069846,
      "balance_loss_clip": 0.99946702,
      "balance_loss_mlp": 1.04910398,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 12.418676149382657,
      "language_loss": 0.6774869,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.7009927,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.9383273124694824
    },
    {
      "auxiliary_loss_clip": 0.01287974,
      "auxiliary_loss_mlp": 0.01058096,
      "balance_loss_clip": 1.00188601,
      "balance_loss_mlp": 1.03715086,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 14.968609139129603,
      "language_loss": 0.7695123,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.79297298,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.6945791244506836
    },
    {
      "auxiliary_loss_clip": 0.01278047,
      "auxiliary_loss_mlp": 0.01053476,
      "balance_loss_clip": 1.03864598,
      "balance_loss_mlp": 1.03238821,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 2.110880932477581,
      "language_loss": 0.80620861,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.8295238,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 2.7216761112213135
    },
    {
      "auxiliary_loss_clip": 0.01286382,
      "auxiliary_loss_mlp": 0.0106203,
      "balance_loss_clip": 1.00254655,
      "balance_loss_mlp": 1.04087043,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.4634344458567385,
      "language_loss": 0.93328261,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95676672,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.8128373622894287
    },
    {
      "auxiliary_loss_clip": 0.01280202,
      "auxiliary_loss_mlp": 0.01132202,
      "balance_loss_clip": 0.96306694,
      "balance_loss_mlp": 0.0,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.5220029447969985,
      "language_loss": 0.75659072,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.78071475,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.906956672668457
    },
    {
      "auxiliary_loss_clip": 0.01272085,
      "auxiliary_loss_mlp": 0.01057177,
      "balance_loss_clip": 0.99756658,
      "balance_loss_mlp": 1.03505182,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 2.368210113097231,
      "language_loss": 0.83597898,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.85927165,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.8525054454803467
    },
    {
      "auxiliary_loss_clip": 0.01285022,
      "auxiliary_loss_mlp": 0.01057754,
      "balance_loss_clip": 1.03642964,
      "balance_loss_mlp": 1.03814363,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 2.2269295103170643,
      "language_loss": 0.85363781,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87706554,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.813359498977661
    },
    {
      "auxiliary_loss_clip": 0.01294074,
      "auxiliary_loss_mlp": 0.01051951,
      "balance_loss_clip": 1.07988214,
      "balance_loss_mlp": 1.03153026,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 1.8508131706216202,
      "language_loss": 0.85516,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.87862021,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.714059829711914
    },
    {
      "auxiliary_loss_clip": 0.01217069,
      "auxiliary_loss_mlp": 0.01127258,
      "balance_loss_clip": 0.96396875,
      "balance_loss_mlp": 0.0,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8744775902311162,
      "language_loss": 0.63108802,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.6545313,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.3359267711639404
    },
    {
      "auxiliary_loss_clip": 0.0120708,
      "auxiliary_loss_mlp": 0.0112716,
      "balance_loss_clip": 1.07158542,
      "balance_loss_mlp": 0.0,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 5.875594642712059,
      "language_loss": 0.59903908,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.62238145,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.177755832672119
    },
    {
      "auxiliary_loss_clip": 0.01205343,
      "auxiliary_loss_mlp": 0.01018706,
      "balance_loss_clip": 0.9891032,
      "balance_loss_mlp": 1.00382853,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.936872519047258,
      "language_loss": 0.63136101,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.65360141,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 3.0225021839141846
    },
    {
      "auxiliary_loss_clip": 0.01282019,
      "auxiliary_loss_mlp": 0.01058183,
      "balance_loss_clip": 0.99748194,
      "balance_loss_mlp": 1.03770304,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 1.8044563733163017,
      "language_loss": 0.79046977,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.81387174,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 2.7928214073181152
    },
    {
      "auxiliary_loss_clip": 0.01202221,
      "auxiliary_loss_mlp": 0.01016603,
      "balance_loss_clip": 1.06768608,
      "balance_loss_mlp": 1.0018214,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8357430563673366,
      "language_loss": 0.5536657,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57585394,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 4.039946556091309
    },
    {
      "auxiliary_loss_clip": 0.01264491,
      "auxiliary_loss_mlp": 0.01064573,
      "balance_loss_clip": 0.873362,
      "balance_loss_mlp": 1.04359198,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.222095152124562,
      "language_loss": 0.78103185,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.80432254,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 5.105177879333496
    },
    {
      "auxiliary_loss_clip": 0.01254354,
      "auxiliary_loss_mlp": 0.01052979,
      "balance_loss_clip": 0.95359534,
      "balance_loss_mlp": 1.03315449,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 2.026826083231853,
      "language_loss": 0.86012304,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.88319635,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 2.9751672744750977
    },
    {
      "auxiliary_loss_clip": 0.01260865,
      "auxiliary_loss_mlp": 0.01058439,
      "balance_loss_clip": 0.88150817,
      "balance_loss_mlp": 1.0382452,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 1.825641433172261,
      "language_loss": 0.93589085,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.95908391,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 4.249449253082275
    },
    {
      "auxiliary_loss_clip": 0.01280464,
      "auxiliary_loss_mlp": 0.01062485,
      "balance_loss_clip": 1.00226784,
      "balance_loss_mlp": 1.04239821,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.772532096349031,
      "language_loss": 0.9148041,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93823361,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 3.107560873031616
    },
    {
      "auxiliary_loss_clip": 0.01284946,
      "auxiliary_loss_mlp": 0.01054869,
      "balance_loss_clip": 1.03863776,
      "balance_loss_mlp": 1.03528321,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 1.9234829149061263,
      "language_loss": 0.80819523,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.83159339,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 2.854412317276001
    },
    {
      "auxiliary_loss_clip": 0.01290995,
      "auxiliary_loss_mlp": 0.01062309,
      "balance_loss_clip": 1.0820874,
      "balance_loss_mlp": 1.04262805,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 2.063037764150344,
      "language_loss": 0.92906946,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.95260251,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 2.823514223098755
    },
    {
      "auxiliary_loss_clip": 0.01201404,
      "auxiliary_loss_mlp": 0.01013897,
      "balance_loss_clip": 0.9882552,
      "balance_loss_mlp": 0.99940091,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8888762789517004,
      "language_loss": 0.59324265,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61539567,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.469059467315674
    },
    {
      "auxiliary_loss_clip": 0.01251087,
      "auxiliary_loss_mlp": 0.01058273,
      "balance_loss_clip": 0.99412256,
      "balance_loss_mlp": 1.03837717,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 2.087182224990165,
      "language_loss": 0.92708457,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.95017827,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.844461441040039
    },
    {
      "auxiliary_loss_clip": 0.01283777,
      "auxiliary_loss_mlp": 0.01063372,
      "balance_loss_clip": 1.03930724,
      "balance_loss_mlp": 1.04311883,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 2.5296827178167294,
      "language_loss": 0.82967877,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85315025,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.772704601287842
    },
    {
      "auxiliary_loss_clip": 0.01279471,
      "auxiliary_loss_mlp": 0.01055304,
      "balance_loss_clip": 1.03957427,
      "balance_loss_mlp": 1.0354439,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 3.2806305612114945,
      "language_loss": 0.73845446,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.7618022,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.855928659439087
    },
    {
      "auxiliary_loss_clip": 0.01273585,
      "auxiliary_loss_mlp": 0.01058779,
      "balance_loss_clip": 1.00129902,
      "balance_loss_mlp": 1.03849018,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 1.9125808172920526,
      "language_loss": 0.84891421,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87223786,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.7593653202056885
    },
    {
      "auxiliary_loss_clip": 0.01286993,
      "auxiliary_loss_mlp": 0.01057844,
      "balance_loss_clip": 1.00420988,
      "balance_loss_mlp": 1.03748286,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 11.753047436629835,
      "language_loss": 0.8154788,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.83892727,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.7007040977478027
    },
    {
      "auxiliary_loss_clip": 0.01287329,
      "auxiliary_loss_mlp": 0.01054496,
      "balance_loss_clip": 1.0388577,
      "balance_loss_mlp": 1.03470683,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 1.9881664305880091,
      "language_loss": 0.78620076,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.80961907,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 2.6770401000976562
    },
    {
      "auxiliary_loss_clip": 0.01290078,
      "auxiliary_loss_mlp": 0.01053975,
      "balance_loss_clip": 1.08019876,
      "balance_loss_mlp": 1.03473473,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 6.765922218901882,
      "language_loss": 0.82069802,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.8441385,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.6923294067382812
    },
    {
      "auxiliary_loss_clip": 0.01288512,
      "auxiliary_loss_mlp": 0.01064763,
      "balance_loss_clip": 1.04421771,
      "balance_loss_mlp": 1.04504609,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 2.228260065815346,
      "language_loss": 0.87041271,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89394546,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.6741702556610107
    },
    {
      "auxiliary_loss_clip": 0.01267468,
      "auxiliary_loss_mlp": 0.01058627,
      "balance_loss_clip": 0.99466896,
      "balance_loss_mlp": 1.0390054,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 3.85073069261129,
      "language_loss": 0.77214986,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79541087,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 2.7355105876922607
    },
    {
      "auxiliary_loss_clip": 0.01280281,
      "auxiliary_loss_mlp": 0.01055098,
      "balance_loss_clip": 1.07530141,
      "balance_loss_mlp": 1.0361315,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 1.942298171554913,
      "language_loss": 0.73986846,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.76322228,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.668098211288452
    },
    {
      "auxiliary_loss_clip": 0.01283909,
      "auxiliary_loss_mlp": 0.01057467,
      "balance_loss_clip": 1.04070878,
      "balance_loss_mlp": 1.03759527,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 3.1636325622341066,
      "language_loss": 0.86666262,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.89007634,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.644425868988037
    },
    {
      "auxiliary_loss_clip": 0.01288556,
      "auxiliary_loss_mlp": 0.01058277,
      "balance_loss_clip": 1.04269886,
      "balance_loss_mlp": 1.03869104,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 2.2367666066246885,
      "language_loss": 0.92615795,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.94962633,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 2.722057819366455
    },
    {
      "auxiliary_loss_clip": 0.01272285,
      "auxiliary_loss_mlp": 0.01061881,
      "balance_loss_clip": 0.9981575,
      "balance_loss_mlp": 1.04256868,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.3848495436885355,
      "language_loss": 0.87585652,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.89919817,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.660248279571533
    },
    {
      "auxiliary_loss_clip": 0.01290037,
      "auxiliary_loss_mlp": 0.01065039,
      "balance_loss_clip": 1.00288153,
      "balance_loss_mlp": 1.04556084,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 3.152791782952014,
      "language_loss": 0.87756157,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.90111232,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.6919097900390625
    },
    {
      "auxiliary_loss_clip": 0.01276051,
      "auxiliary_loss_mlp": 0.0106848,
      "balance_loss_clip": 0.99788499,
      "balance_loss_mlp": 1.04914403,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 1.8366203002459167,
      "language_loss": 0.71099269,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73443806,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 2.9438095092773438
    },
    {
      "auxiliary_loss_clip": 0.01282552,
      "auxiliary_loss_mlp": 0.01063799,
      "balance_loss_clip": 1.00404406,
      "balance_loss_mlp": 1.04430819,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 2.3754700399774937,
      "language_loss": 0.83885193,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86231554,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.7110438346862793
    },
    {
      "auxiliary_loss_clip": 0.01273765,
      "auxiliary_loss_mlp": 0.01060129,
      "balance_loss_clip": 0.84664518,
      "balance_loss_mlp": 1.03966093,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.5196639186893413,
      "language_loss": 0.81028104,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.83362007,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 3.062098979949951
    },
    {
      "auxiliary_loss_clip": 0.0127652,
      "auxiliary_loss_mlp": 0.01055692,
      "balance_loss_clip": 0.92049766,
      "balance_loss_mlp": 1.03601098,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 2.907231545118109,
      "language_loss": 0.9348343,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95815641,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 3.39823055267334
    },
    {
      "auxiliary_loss_clip": 0.01245672,
      "auxiliary_loss_mlp": 0.01049709,
      "balance_loss_clip": 0.95487475,
      "balance_loss_mlp": 1.03062367,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.7550854246504093,
      "language_loss": 0.8644622,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.88741601,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 3.7952721118927
    },
    {
      "auxiliary_loss_clip": 0.01279831,
      "auxiliary_loss_mlp": 0.01054422,
      "balance_loss_clip": 0.9951759,
      "balance_loss_mlp": 1.03520536,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 1.9178615541084696,
      "language_loss": 0.80070722,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82404971,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 4.788560390472412
    },
    {
      "auxiliary_loss_clip": 0.01288511,
      "auxiliary_loss_mlp": 0.01067799,
      "balance_loss_clip": 1.07844186,
      "balance_loss_mlp": 1.04773629,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 1.9618555421730064,
      "language_loss": 0.87173527,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89529836,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 2.6503851413726807
    },
    {
      "auxiliary_loss_clip": 0.01280146,
      "auxiliary_loss_mlp": 0.01060331,
      "balance_loss_clip": 0.99989271,
      "balance_loss_mlp": 1.04038763,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 2.072257675760028,
      "language_loss": 0.82486057,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.84826535,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 3.7878775596618652
    },
    {
      "auxiliary_loss_clip": 0.01285606,
      "auxiliary_loss_mlp": 0.01056737,
      "balance_loss_clip": 0.99664444,
      "balance_loss_mlp": 1.03749704,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 1.5703606910322003,
      "language_loss": 0.80161339,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82503688,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 2.83009934425354
    },
    {
      "auxiliary_loss_clip": 0.01268531,
      "auxiliary_loss_mlp": 0.01051642,
      "balance_loss_clip": 0.92241812,
      "balance_loss_mlp": 1.03202057,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 2.0962684569022683,
      "language_loss": 0.84256768,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86576939,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 2.8554723262786865
    },
    {
      "auxiliary_loss_clip": 0.01285598,
      "auxiliary_loss_mlp": 0.01061727,
      "balance_loss_clip": 1.07894707,
      "balance_loss_mlp": 1.04254675,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 3.9812703386256985,
      "language_loss": 0.89638519,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.91985846,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.724647045135498
    },
    {
      "auxiliary_loss_clip": 0.01287843,
      "auxiliary_loss_mlp": 0.01055938,
      "balance_loss_clip": 1.03895986,
      "balance_loss_mlp": 1.0365901,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 2.0726171731471625,
      "language_loss": 0.88289297,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.90633082,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.7103519439697266
    },
    {
      "auxiliary_loss_clip": 0.0126466,
      "auxiliary_loss_mlp": 0.01057043,
      "balance_loss_clip": 0.95822722,
      "balance_loss_mlp": 1.03804123,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.412055359341681,
      "language_loss": 0.81268227,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83589935,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.8073015213012695
    },
    {
      "auxiliary_loss_clip": 0.01277355,
      "auxiliary_loss_mlp": 0.0105057,
      "balance_loss_clip": 0.96023226,
      "balance_loss_mlp": 1.03130615,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 2.3748796463471966,
      "language_loss": 0.84984124,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.87312055,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.730539083480835
    },
    {
      "auxiliary_loss_clip": 0.01278356,
      "auxiliary_loss_mlp": 0.01047184,
      "balance_loss_clip": 0.99966079,
      "balance_loss_mlp": 1.02724051,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 1.6178371456149192,
      "language_loss": 0.80608428,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.82933974,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.76692271232605
    },
    {
      "auxiliary_loss_clip": 0.0126284,
      "auxiliary_loss_mlp": 0.01058213,
      "balance_loss_clip": 0.99846864,
      "balance_loss_mlp": 1.03904438,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 1.9346640876581562,
      "language_loss": 0.88001215,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90322274,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.753382682800293
    },
    {
      "auxiliary_loss_clip": 0.0126569,
      "auxiliary_loss_mlp": 0.01051517,
      "balance_loss_clip": 0.95991492,
      "balance_loss_mlp": 1.03278899,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 1.8529556074196931,
      "language_loss": 0.85998523,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88315725,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.8550143241882324
    },
    {
      "auxiliary_loss_clip": 0.01286434,
      "auxiliary_loss_mlp": 0.01131664,
      "balance_loss_clip": 1.08201241,
      "balance_loss_mlp": 0.0,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 2.3824550490453613,
      "language_loss": 0.82565755,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.8498385,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.760143995285034
    },
    {
      "auxiliary_loss_clip": 0.01284354,
      "auxiliary_loss_mlp": 0.01050412,
      "balance_loss_clip": 1.04004729,
      "balance_loss_mlp": 1.03043211,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.7328413411741677,
      "language_loss": 0.84154499,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.8648926,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 2.782965660095215
    },
    {
      "auxiliary_loss_clip": 0.01287499,
      "auxiliary_loss_mlp": 0.01050989,
      "balance_loss_clip": 1.07969761,
      "balance_loss_mlp": 1.03215361,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 2.334270797157165,
      "language_loss": 0.80579847,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.82918334,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 2.6583127975463867
    },
    {
      "auxiliary_loss_clip": 0.01253818,
      "auxiliary_loss_mlp": 0.01056142,
      "balance_loss_clip": 0.95545721,
      "balance_loss_mlp": 1.03723586,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 2.2728986310284185,
      "language_loss": 0.86710984,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.89020944,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.7603979110717773
    },
    {
      "auxiliary_loss_clip": 0.01292851,
      "auxiliary_loss_mlp": 0.01054529,
      "balance_loss_clip": 1.08487153,
      "balance_loss_mlp": 1.03588414,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 2.251959397518022,
      "language_loss": 0.76876926,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.792243,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.7168002128601074
    },
    {
      "auxiliary_loss_clip": 0.0127897,
      "auxiliary_loss_mlp": 0.01061814,
      "balance_loss_clip": 1.000687,
      "balance_loss_mlp": 1.04282379,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 2.020271755552471,
      "language_loss": 0.90527058,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.92867845,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 3.0527892112731934
    },
    {
      "auxiliary_loss_clip": 0.01271133,
      "auxiliary_loss_mlp": 0.0113177,
      "balance_loss_clip": 0.99688792,
      "balance_loss_mlp": 0.0,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 1.8601790938023324,
      "language_loss": 0.82253587,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84656489,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 2.9431068897247314
    },
    {
      "auxiliary_loss_clip": 0.01203514,
      "auxiliary_loss_mlp": 0.01028642,
      "balance_loss_clip": 1.03102541,
      "balance_loss_mlp": 1.01471806,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.8406451608896108,
      "language_loss": 0.55108654,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57340813,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.3525002002716064
    },
    {
      "auxiliary_loss_clip": 0.01274684,
      "auxiliary_loss_mlp": 0.01064862,
      "balance_loss_clip": 0.95858425,
      "balance_loss_mlp": 1.0454433,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 2.3330090281602796,
      "language_loss": 0.7872104,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.81060588,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.86265230178833
    },
    {
      "auxiliary_loss_clip": 0.01283728,
      "auxiliary_loss_mlp": 0.0106999,
      "balance_loss_clip": 1.07770729,
      "balance_loss_mlp": 1.05045152,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 1.8638357867590605,
      "language_loss": 0.81012464,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.83366185,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.767510175704956
    },
    {
      "auxiliary_loss_clip": 0.01288419,
      "auxiliary_loss_mlp": 0.01057741,
      "balance_loss_clip": 1.03985643,
      "balance_loss_mlp": 1.03856039,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.1460848586991017,
      "language_loss": 0.77414113,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.79760265,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.688371181488037
    },
    {
      "auxiliary_loss_clip": 0.0127669,
      "auxiliary_loss_mlp": 0.01065733,
      "balance_loss_clip": 0.99720091,
      "balance_loss_mlp": 1.0462184,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 2.331880406505502,
      "language_loss": 0.87626165,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.89968586,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.727959394454956
    },
    {
      "auxiliary_loss_clip": 0.0128617,
      "auxiliary_loss_mlp": 0.01062219,
      "balance_loss_clip": 1.04011273,
      "balance_loss_mlp": 1.04305053,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 1.8752180947351422,
      "language_loss": 0.7799629,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80344677,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.80051326751709
    },
    {
      "auxiliary_loss_clip": 0.0126267,
      "auxiliary_loss_mlp": 0.01058793,
      "balance_loss_clip": 0.95749295,
      "balance_loss_mlp": 1.03778815,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 3.5188512184593175,
      "language_loss": 0.87939656,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.9026112,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 2.7413275241851807
    },
    {
      "auxiliary_loss_clip": 0.01277761,
      "auxiliary_loss_mlp": 0.01050355,
      "balance_loss_clip": 0.99695164,
      "balance_loss_mlp": 1.03156793,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 2.7477193416088315,
      "language_loss": 0.87828457,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.90156579,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 4.635085821151733
    },
    {
      "auxiliary_loss_clip": 0.01281861,
      "auxiliary_loss_mlp": 0.01051876,
      "balance_loss_clip": 1.0403825,
      "balance_loss_mlp": 1.03302908,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 3.086160482744592,
      "language_loss": 0.81027031,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.83360767,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 3.6199405193328857
    },
    {
      "auxiliary_loss_clip": 0.01257179,
      "auxiliary_loss_mlp": 0.01066197,
      "balance_loss_clip": 0.99272293,
      "balance_loss_mlp": 1.04742193,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.3594426297838527,
      "language_loss": 0.77782035,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.80105412,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 3.826721429824829
    },
    {
      "auxiliary_loss_clip": 0.01276704,
      "auxiliary_loss_mlp": 0.01131914,
      "balance_loss_clip": 1.03577709,
      "balance_loss_mlp": 0.0,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 2.1295415894850236,
      "language_loss": 0.81785655,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.84194273,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 2.7208027839660645
    },
    {
      "auxiliary_loss_clip": 0.01276847,
      "auxiliary_loss_mlp": 0.01051054,
      "balance_loss_clip": 0.9965502,
      "balance_loss_mlp": 1.03159904,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 3.272914366584611,
      "language_loss": 0.87908554,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90236455,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 2.748518466949463
    },
    {
      "auxiliary_loss_clip": 0.0128321,
      "auxiliary_loss_mlp": 0.01050515,
      "balance_loss_clip": 1.07818103,
      "balance_loss_mlp": 1.0318352,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 1.943115260728911,
      "language_loss": 0.81269288,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.83603013,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.694455623626709
    },
    {
      "auxiliary_loss_clip": 0.01285904,
      "auxiliary_loss_mlp": 0.01058178,
      "balance_loss_clip": 1.07802582,
      "balance_loss_mlp": 1.03894997,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 3.010366875135687,
      "language_loss": 0.81037962,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83382046,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.6747994422912598
    },
    {
      "auxiliary_loss_clip": 0.01278182,
      "auxiliary_loss_mlp": 0.01050071,
      "balance_loss_clip": 1.03557014,
      "balance_loss_mlp": 1.02979374,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.267805367007421,
      "language_loss": 0.76658738,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.78986996,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.6775646209716797
    },
    {
      "auxiliary_loss_clip": 0.01279573,
      "auxiliary_loss_mlp": 0.01063183,
      "balance_loss_clip": 1.0387013,
      "balance_loss_mlp": 1.0430243,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 2.193503603368784,
      "language_loss": 0.7864821,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.80990964,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.7757959365844727
    },
    {
      "auxiliary_loss_clip": 0.01270882,
      "auxiliary_loss_mlp": 0.01132118,
      "balance_loss_clip": 0.95495307,
      "balance_loss_mlp": 0.0,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 1.9379328795870006,
      "language_loss": 0.90678549,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.93081558,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.7437403202056885
    },
    {
      "auxiliary_loss_clip": 0.01262293,
      "auxiliary_loss_mlp": 0.01057195,
      "balance_loss_clip": 1.03276896,
      "balance_loss_mlp": 1.03816915,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.2845317770526012,
      "language_loss": 0.93289876,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95609355,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.7001490592956543
    },
    {
      "auxiliary_loss_clip": 0.01285025,
      "auxiliary_loss_mlp": 0.01038578,
      "balance_loss_clip": 1.07892907,
      "balance_loss_mlp": 1.01988566,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 1.9167489419175017,
      "language_loss": 0.78112042,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80435646,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.723466157913208
    },
    {
      "auxiliary_loss_clip": 0.01269427,
      "auxiliary_loss_mlp": 0.01056917,
      "balance_loss_clip": 0.96327353,
      "balance_loss_mlp": 1.03660393,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 1.8894253783127366,
      "language_loss": 0.81156731,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83483076,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.7631378173828125
    },
    {
      "auxiliary_loss_clip": 0.01268069,
      "auxiliary_loss_mlp": 0.01047796,
      "balance_loss_clip": 0.91844475,
      "balance_loss_mlp": 1.0289731,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.3854529174528376,
      "language_loss": 0.84695786,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.87011653,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.736171245574951
    },
    {
      "auxiliary_loss_clip": 0.01276268,
      "auxiliary_loss_mlp": 0.01131453,
      "balance_loss_clip": 0.99991918,
      "balance_loss_mlp": 0.0,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 1.6604436269932343,
      "language_loss": 0.9049753,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.92905247,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 2.7023308277130127
    },
    {
      "auxiliary_loss_clip": 0.01271712,
      "auxiliary_loss_mlp": 0.01050962,
      "balance_loss_clip": 0.95531154,
      "balance_loss_mlp": 1.03079212,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.470127922383296,
      "language_loss": 0.93880343,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96203017,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.7062010765075684
    },
    {
      "auxiliary_loss_clip": 0.01279723,
      "auxiliary_loss_mlp": 0.01063959,
      "balance_loss_clip": 1.00225151,
      "balance_loss_mlp": 1.04479027,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 2.03539939871987,
      "language_loss": 0.71017963,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73361647,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 2.648792028427124
    },
    {
      "auxiliary_loss_clip": 0.01284458,
      "auxiliary_loss_mlp": 0.01131683,
      "balance_loss_clip": 1.04154551,
      "balance_loss_mlp": 0.0,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 2.0481192586115275,
      "language_loss": 0.78559858,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.80975997,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.6969640254974365
    },
    {
      "auxiliary_loss_clip": 0.01254814,
      "auxiliary_loss_mlp": 0.01054167,
      "balance_loss_clip": 0.99069679,
      "balance_loss_mlp": 1.03540373,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 2.571494418419231,
      "language_loss": 0.74112177,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.76421154,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 2.6279942989349365
    },
    {
      "auxiliary_loss_clip": 0.01259731,
      "auxiliary_loss_mlp": 0.01051801,
      "balance_loss_clip": 1.03105772,
      "balance_loss_mlp": 1.03312063,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 2.873482339334522,
      "language_loss": 0.91554391,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.93865925,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 2.703270196914673
    },
    {
      "auxiliary_loss_clip": 0.01261346,
      "auxiliary_loss_mlp": 0.01046162,
      "balance_loss_clip": 1.03143907,
      "balance_loss_mlp": 1.02741075,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 1.7468184933794686,
      "language_loss": 0.67797267,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70104778,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.6488122940063477
    },
    {
      "auxiliary_loss_clip": 0.01252567,
      "auxiliary_loss_mlp": 0.0105611,
      "balance_loss_clip": 0.95663768,
      "balance_loss_mlp": 1.03733468,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.792277880000664,
      "language_loss": 0.94882923,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97191596,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.7236690521240234
    },
    {
      "auxiliary_loss_clip": 0.01274949,
      "auxiliary_loss_mlp": 0.01046753,
      "balance_loss_clip": 0.96100748,
      "balance_loss_mlp": 1.02763224,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 2.2923910342863283,
      "language_loss": 0.74738884,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77060592,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.8186583518981934
    },
    {
      "auxiliary_loss_clip": 0.01281792,
      "auxiliary_loss_mlp": 0.01057755,
      "balance_loss_clip": 1.04124355,
      "balance_loss_mlp": 1.03821659,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 7.9755203698380495,
      "language_loss": 0.80971313,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.8331086,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.72200345993042
    },
    {
      "auxiliary_loss_clip": 0.01274951,
      "auxiliary_loss_mlp": 0.01059654,
      "balance_loss_clip": 1.04243469,
      "balance_loss_mlp": 1.04083109,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 2.112161783143218,
      "language_loss": 0.80097556,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82432157,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 2.7261440753936768
    },
    {
      "auxiliary_loss_clip": 0.01279843,
      "auxiliary_loss_mlp": 0.01058778,
      "balance_loss_clip": 1.04237604,
      "balance_loss_mlp": 1.03984809,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 3.0142453660137707,
      "language_loss": 0.93749958,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96088576,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 3.6707425117492676
    },
    {
      "auxiliary_loss_clip": 0.01256556,
      "auxiliary_loss_mlp": 0.0105604,
      "balance_loss_clip": 0.92350674,
      "balance_loss_mlp": 1.03693068,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 2.1068346404294456,
      "language_loss": 0.82800412,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85113007,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 4.6387481689453125
    },
    {
      "auxiliary_loss_clip": 0.01234211,
      "auxiliary_loss_mlp": 0.01052964,
      "balance_loss_clip": 0.87111652,
      "balance_loss_mlp": 1.03337824,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 2.345085022584433,
      "language_loss": 0.80318236,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.8260541,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 2.902627944946289
    },
    {
      "auxiliary_loss_clip": 0.01281468,
      "auxiliary_loss_mlp": 0.01056247,
      "balance_loss_clip": 1.03663647,
      "balance_loss_mlp": 1.03654182,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 2.4702459244756243,
      "language_loss": 0.80274349,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.82612062,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 3.729170560836792
    },
    {
      "auxiliary_loss_clip": 0.01281956,
      "auxiliary_loss_mlp": 0.0105468,
      "balance_loss_clip": 1.07578242,
      "balance_loss_mlp": 1.03541529,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 2.0256878272231265,
      "language_loss": 0.86454415,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.8879106,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 2.5740036964416504
    },
    {
      "auxiliary_loss_clip": 0.01264674,
      "auxiliary_loss_mlp": 0.01057771,
      "balance_loss_clip": 0.99592185,
      "balance_loss_mlp": 1.03791082,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 2.124715361557393,
      "language_loss": 0.76693928,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.79016376,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 2.6528217792510986
    },
    {
      "auxiliary_loss_clip": 0.0127572,
      "auxiliary_loss_mlp": 0.01058851,
      "balance_loss_clip": 1.00316978,
      "balance_loss_mlp": 1.03933668,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.1098291220312273,
      "language_loss": 0.84551919,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.86886489,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.6632275581359863
    },
    {
      "auxiliary_loss_clip": 0.01280283,
      "auxiliary_loss_mlp": 0.01057537,
      "balance_loss_clip": 1.03536725,
      "balance_loss_mlp": 1.03853452,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.176624623840006,
      "language_loss": 0.94645107,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.96982926,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.6655571460723877
    },
    {
      "auxiliary_loss_clip": 0.01273829,
      "auxiliary_loss_mlp": 0.01056987,
      "balance_loss_clip": 0.99979645,
      "balance_loss_mlp": 1.03768671,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 1.909211867609446,
      "language_loss": 0.8675456,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.8908537,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.677464485168457
    },
    {
      "auxiliary_loss_clip": 0.01274485,
      "auxiliary_loss_mlp": 0.01049778,
      "balance_loss_clip": 0.96077573,
      "balance_loss_mlp": 1.03034651,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 3.230855715869339,
      "language_loss": 0.73550886,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.75875151,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.8531460762023926
    },
    {
      "auxiliary_loss_clip": 0.01273767,
      "auxiliary_loss_mlp": 0.01057811,
      "balance_loss_clip": 0.99728054,
      "balance_loss_mlp": 1.0376054,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.3252545460340968,
      "language_loss": 0.78787816,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81119394,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.7476143836975098
    },
    {
      "auxiliary_loss_clip": 0.01281658,
      "auxiliary_loss_mlp": 0.01057153,
      "balance_loss_clip": 1.07846451,
      "balance_loss_mlp": 1.03798461,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 2.1294095901862025,
      "language_loss": 0.87443721,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.89782536,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.6723520755767822
    },
    {
      "auxiliary_loss_clip": 0.01285121,
      "auxiliary_loss_mlp": 0.0105774,
      "balance_loss_clip": 1.04459357,
      "balance_loss_mlp": 1.0391438,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 2.2168488352674136,
      "language_loss": 0.79736221,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.82079089,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.6518588066101074
    },
    {
      "auxiliary_loss_clip": 0.01277456,
      "auxiliary_loss_mlp": 0.01060154,
      "balance_loss_clip": 1.07794356,
      "balance_loss_mlp": 1.04056764,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 2.122504359350591,
      "language_loss": 0.91595417,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.93933028,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.6924757957458496
    },
    {
      "auxiliary_loss_clip": 0.01275996,
      "auxiliary_loss_mlp": 0.01049369,
      "balance_loss_clip": 1.03977132,
      "balance_loss_mlp": 1.03003335,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.2265329985639513,
      "language_loss": 0.88030422,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.90355784,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.713550329208374
    },
    {
      "auxiliary_loss_clip": 0.0127586,
      "auxiliary_loss_mlp": 0.01051265,
      "balance_loss_clip": 1.04067862,
      "balance_loss_mlp": 1.03190541,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.1905602930763566,
      "language_loss": 0.83840597,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86167723,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.700380802154541
    },
    {
      "auxiliary_loss_clip": 0.01276298,
      "auxiliary_loss_mlp": 0.01043576,
      "balance_loss_clip": 1.03983355,
      "balance_loss_mlp": 1.0249083,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.098450209863753,
      "language_loss": 0.76441294,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.78761172,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 2.649432420730591
    },
    {
      "auxiliary_loss_clip": 0.01274562,
      "auxiliary_loss_mlp": 0.01052177,
      "balance_loss_clip": 1.03983831,
      "balance_loss_mlp": 1.03297257,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.282833176071588,
      "language_loss": 0.90441424,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.92768162,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.6621100902557373
    },
    {
      "auxiliary_loss_clip": 0.0126677,
      "auxiliary_loss_mlp": 0.0106024,
      "balance_loss_clip": 0.95634341,
      "balance_loss_mlp": 1.04113066,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.9612274333375461,
      "language_loss": 0.85335904,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87662911,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 2.697397232055664
    },
    {
      "auxiliary_loss_clip": 0.01259491,
      "auxiliary_loss_mlp": 0.01132006,
      "balance_loss_clip": 0.95692277,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.308663529522769,
      "language_loss": 0.89439023,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.91830528,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.680802822113037
    },
    {
      "auxiliary_loss_clip": 0.01278869,
      "auxiliary_loss_mlp": 0.01058405,
      "balance_loss_clip": 1.04000556,
      "balance_loss_mlp": 1.03899741,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 2.113047774960346,
      "language_loss": 0.88305807,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.90643084,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 2.7365224361419678
    },
    {
      "auxiliary_loss_clip": 0.01267202,
      "auxiliary_loss_mlp": 0.01062358,
      "balance_loss_clip": 1.03632641,
      "balance_loss_mlp": 1.04339194,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 2.0530113391745237,
      "language_loss": 0.76766348,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.79095918,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.660155773162842
    },
    {
      "auxiliary_loss_clip": 0.01275059,
      "auxiliary_loss_mlp": 0.01047262,
      "balance_loss_clip": 0.96018094,
      "balance_loss_mlp": 1.02932119,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 2.5332070375220113,
      "language_loss": 0.76568508,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.7889083,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.783735513687134
    },
    {
      "auxiliary_loss_clip": 0.01278577,
      "auxiliary_loss_mlp": 0.01055191,
      "balance_loss_clip": 1.07803082,
      "balance_loss_mlp": 1.03663015,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 1.638979312110473,
      "language_loss": 0.80933321,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83267093,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.7125532627105713
    },
    {
      "auxiliary_loss_clip": 0.01270601,
      "auxiliary_loss_mlp": 0.0106186,
      "balance_loss_clip": 0.99616557,
      "balance_loss_mlp": 1.04271531,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.8913533640874576,
      "language_loss": 0.85381007,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.87713468,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.703601598739624
    },
    {
      "auxiliary_loss_clip": 0.01274843,
      "auxiliary_loss_mlp": 0.01055097,
      "balance_loss_clip": 1.03924429,
      "balance_loss_mlp": 1.03483188,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.0963488985434537,
      "language_loss": 0.77770376,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.80100322,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 2.647484302520752
    },
    {
      "auxiliary_loss_clip": 0.0127055,
      "auxiliary_loss_mlp": 0.01055255,
      "balance_loss_clip": 1.00103998,
      "balance_loss_mlp": 1.03619361,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 2.5631178877244554,
      "language_loss": 0.96346414,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98672211,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 2.687753200531006
    },
    {
      "auxiliary_loss_clip": 0.01265161,
      "auxiliary_loss_mlp": 0.01052719,
      "balance_loss_clip": 0.9591713,
      "balance_loss_mlp": 1.03482568,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 2.122900419592608,
      "language_loss": 0.87995064,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90312946,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 5.4813220500946045
    },
    {
      "auxiliary_loss_clip": 0.01273618,
      "auxiliary_loss_mlp": 0.01054131,
      "balance_loss_clip": 1.00029325,
      "balance_loss_mlp": 1.03518856,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.690698524510862,
      "language_loss": 0.82531697,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84859443,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 2.7343897819519043
    },
    {
      "auxiliary_loss_clip": 0.01261564,
      "auxiliary_loss_mlp": 0.01066856,
      "balance_loss_clip": 0.95581239,
      "balance_loss_mlp": 1.04778242,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 2.668717955668562,
      "language_loss": 0.70487463,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72815883,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 2.7462780475616455
    },
    {
      "auxiliary_loss_clip": 0.01277892,
      "auxiliary_loss_mlp": 0.01131315,
      "balance_loss_clip": 1.03843963,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 2.1484646099406133,
      "language_loss": 0.95484716,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.97893918,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 3.654343366622925
    },
    {
      "auxiliary_loss_clip": 0.01266925,
      "auxiliary_loss_mlp": 0.01059534,
      "balance_loss_clip": 0.96420902,
      "balance_loss_mlp": 1.04117608,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 1.7714274681691795,
      "language_loss": 0.91442716,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93769181,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 2.7294673919677734
    },
    {
      "auxiliary_loss_clip": 0.01272321,
      "auxiliary_loss_mlp": 0.01046567,
      "balance_loss_clip": 1.03380203,
      "balance_loss_mlp": 1.02830434,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 1.924220610605731,
      "language_loss": 0.93241823,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95560718,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.7034099102020264
    },
    {
      "auxiliary_loss_clip": 0.01272213,
      "auxiliary_loss_mlp": 0.01059491,
      "balance_loss_clip": 0.99630004,
      "balance_loss_mlp": 1.04017854,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 2.7792942635942803,
      "language_loss": 0.79875314,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82207012,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.6954257488250732
    },
    {
      "auxiliary_loss_clip": 0.01273445,
      "auxiliary_loss_mlp": 0.01054804,
      "balance_loss_clip": 0.99554408,
      "balance_loss_mlp": 1.0354805,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 1.9595664927496117,
      "language_loss": 0.81360817,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83689064,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.7684597969055176
    },
    {
      "auxiliary_loss_clip": 0.01272852,
      "auxiliary_loss_mlp": 0.01057854,
      "balance_loss_clip": 1.03790534,
      "balance_loss_mlp": 1.03962708,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 1.8245929045752847,
      "language_loss": 0.87145901,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.89476609,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.6678011417388916
    },
    {
      "auxiliary_loss_clip": 0.01266226,
      "auxiliary_loss_mlp": 0.01062598,
      "balance_loss_clip": 0.92113721,
      "balance_loss_mlp": 1.04459763,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.46642589572149,
      "language_loss": 0.80786037,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.83114862,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.779505729675293
    },
    {
      "auxiliary_loss_clip": 0.01254404,
      "auxiliary_loss_mlp": 0.01055645,
      "balance_loss_clip": 0.99602407,
      "balance_loss_mlp": 1.03689384,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 1.6606405912786484,
      "language_loss": 0.91061419,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93371463,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.7766547203063965
    },
    {
      "auxiliary_loss_clip": 0.01265424,
      "auxiliary_loss_mlp": 0.01066172,
      "balance_loss_clip": 0.99678028,
      "balance_loss_mlp": 1.04824305,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 4.341202921219288,
      "language_loss": 0.78653961,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.80985558,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.6609175205230713
    },
    {
      "auxiliary_loss_clip": 0.01259502,
      "auxiliary_loss_mlp": 0.01131596,
      "balance_loss_clip": 0.99305201,
      "balance_loss_mlp": 0.0,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 3.4614729289014186,
      "language_loss": 0.81707644,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.84098744,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.7145276069641113
    },
    {
      "auxiliary_loss_clip": 0.01276413,
      "auxiliary_loss_mlp": 0.0105936,
      "balance_loss_clip": 1.07827437,
      "balance_loss_mlp": 1.0418725,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 2.1610577539357823,
      "language_loss": 0.81492537,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.83828306,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.6236865520477295
    },
    {
      "auxiliary_loss_clip": 0.01273803,
      "auxiliary_loss_mlp": 0.01131982,
      "balance_loss_clip": 1.03561473,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 1.9577633708620472,
      "language_loss": 0.74641258,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.77047038,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.704913377761841
    },
    {
      "auxiliary_loss_clip": 0.01271584,
      "auxiliary_loss_mlp": 0.01061225,
      "balance_loss_clip": 0.99772787,
      "balance_loss_mlp": 1.04323626,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.062349418769638,
      "language_loss": 0.93747723,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96080524,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 2.7048192024230957
    },
    {
      "auxiliary_loss_clip": 0.01268972,
      "auxiliary_loss_mlp": 0.01049413,
      "balance_loss_clip": 1.03696573,
      "balance_loss_mlp": 1.03190148,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 2.592974340006,
      "language_loss": 0.69179541,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.71497929,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 2.972572088241577
    },
    {
      "auxiliary_loss_clip": 0.01270718,
      "auxiliary_loss_mlp": 0.01051646,
      "balance_loss_clip": 0.95898867,
      "balance_loss_mlp": 1.03246582,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 2.7008785486443454,
      "language_loss": 0.76017147,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.78339511,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.7524077892303467
    },
    {
      "auxiliary_loss_clip": 0.01270371,
      "auxiliary_loss_mlp": 0.01044654,
      "balance_loss_clip": 1.03318751,
      "balance_loss_mlp": 1.02615285,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 1.8060514223799562,
      "language_loss": 0.92221379,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.945364,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.7282683849334717
    },
    {
      "auxiliary_loss_clip": 0.01199405,
      "auxiliary_loss_mlp": 0.01016499,
      "balance_loss_clip": 0.99291122,
      "balance_loss_mlp": 1.00343406,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8471440201801419,
      "language_loss": 0.62984955,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65200859,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.3663899898529053
    },
    {
      "auxiliary_loss_clip": 0.01271397,
      "auxiliary_loss_mlp": 0.0105224,
      "balance_loss_clip": 0.96103847,
      "balance_loss_mlp": 1.03432262,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.133844573517153,
      "language_loss": 0.86358899,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88682532,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 2.6734445095062256
    },
    {
      "auxiliary_loss_clip": 0.01276563,
      "auxiliary_loss_mlp": 0.01050487,
      "balance_loss_clip": 1.07468557,
      "balance_loss_mlp": 1.03195012,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 1.7230957008680543,
      "language_loss": 0.74649996,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.7697705,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.6941797733306885
    },
    {
      "auxiliary_loss_clip": 0.01280297,
      "auxiliary_loss_mlp": 0.0105687,
      "balance_loss_clip": 1.07776129,
      "balance_loss_mlp": 1.03745079,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 2.106075874080189,
      "language_loss": 0.70950603,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.73287773,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.652513027191162
    },
    {
      "auxiliary_loss_clip": 0.01265225,
      "auxiliary_loss_mlp": 0.01053041,
      "balance_loss_clip": 0.87631953,
      "balance_loss_mlp": 1.03456354,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 1.9481619545909883,
      "language_loss": 0.77670801,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.7998907,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.9320785999298096
    },
    {
      "auxiliary_loss_clip": 0.01264613,
      "auxiliary_loss_mlp": 0.01054448,
      "balance_loss_clip": 0.92177546,
      "balance_loss_mlp": 1.03593493,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 1.7578487542944237,
      "language_loss": 0.80322754,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82641816,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 2.9694786071777344
    },
    {
      "auxiliary_loss_clip": 0.01270941,
      "auxiliary_loss_mlp": 0.01057132,
      "balance_loss_clip": 0.99721789,
      "balance_loss_mlp": 1.0387615,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 1.936552984724245,
      "language_loss": 0.77663386,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.7999146,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 3.5148019790649414
    },
    {
      "auxiliary_loss_clip": 0.01191273,
      "auxiliary_loss_mlp": 0.0101508,
      "balance_loss_clip": 1.06732583,
      "balance_loss_mlp": 1.00220537,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8857372058128475,
      "language_loss": 0.63866729,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.66073084,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 4.049649715423584
    },
    {
      "auxiliary_loss_clip": 0.01276948,
      "auxiliary_loss_mlp": 0.01057752,
      "balance_loss_clip": 1.07726073,
      "balance_loss_mlp": 1.03852367,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 2.250982452401852,
      "language_loss": 0.80431885,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.82766581,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 3.6034345626831055
    },
    {
      "auxiliary_loss_clip": 0.01268162,
      "auxiliary_loss_mlp": 0.01131237,
      "balance_loss_clip": 1.03640747,
      "balance_loss_mlp": 0.0,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 1.7542902757677798,
      "language_loss": 0.82931501,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85330898,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 2.679945945739746
    },
    {
      "auxiliary_loss_clip": 0.01274382,
      "auxiliary_loss_mlp": 0.01059005,
      "balance_loss_clip": 1.03535175,
      "balance_loss_mlp": 1.04090905,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 8.042297312734624,
      "language_loss": 0.88222289,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90555674,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 3.655757188796997
    },
    {
      "auxiliary_loss_clip": 0.0127001,
      "auxiliary_loss_mlp": 0.01051368,
      "balance_loss_clip": 1.03664637,
      "balance_loss_mlp": 1.03317654,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 2.3527824218360562,
      "language_loss": 0.78751612,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.81072986,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 2.707244396209717
    },
    {
      "auxiliary_loss_clip": 0.01268892,
      "auxiliary_loss_mlp": 0.01131091,
      "balance_loss_clip": 0.99909163,
      "balance_loss_mlp": 0.0,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 1.851809853413032,
      "language_loss": 0.80626488,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.83026469,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.7857816219329834
    },
    {
      "auxiliary_loss_clip": 0.01276136,
      "auxiliary_loss_mlp": 0.01048699,
      "balance_loss_clip": 1.04361701,
      "balance_loss_mlp": 1.03097224,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 6.72311540479711,
      "language_loss": 0.8265155,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.84976387,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.754481792449951
    },
    {
      "auxiliary_loss_clip": 0.01254191,
      "auxiliary_loss_mlp": 0.01059974,
      "balance_loss_clip": 0.91791427,
      "balance_loss_mlp": 1.04200912,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 2.152601226046298,
      "language_loss": 0.66953433,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.69267601,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.80000901222229
    },
    {
      "auxiliary_loss_clip": 0.01275428,
      "auxiliary_loss_mlp": 0.01063487,
      "balance_loss_clip": 1.07543123,
      "balance_loss_mlp": 1.0456177,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 2.3219446792836154,
      "language_loss": 0.85218322,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.87557232,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.7053534984588623
    },
    {
      "auxiliary_loss_clip": 0.01276416,
      "auxiliary_loss_mlp": 0.01051318,
      "balance_loss_clip": 1.07775617,
      "balance_loss_mlp": 1.03312612,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 3.929476574943472,
      "language_loss": 0.84374988,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.86702716,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.679130792617798
    },
    {
      "auxiliary_loss_clip": 0.01242021,
      "auxiliary_loss_mlp": 0.01053384,
      "balance_loss_clip": 0.87553948,
      "balance_loss_mlp": 1.03550243,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 1.9030850153597947,
      "language_loss": 0.8920331,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91498709,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.818049192428589
    },
    {
      "auxiliary_loss_clip": 0.0118524,
      "auxiliary_loss_mlp": 0.01014676,
      "balance_loss_clip": 1.06363583,
      "balance_loss_mlp": 1.00199246,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8211551476710423,
      "language_loss": 0.5709604,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59295952,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.1470839977264404
    },
    {
      "auxiliary_loss_clip": 0.01269864,
      "auxiliary_loss_mlp": 0.01065961,
      "balance_loss_clip": 1.03294492,
      "balance_loss_mlp": 1.0477581,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.430859364416376,
      "language_loss": 0.88435328,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90771151,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.711653470993042
    },
    {
      "auxiliary_loss_clip": 0.01270799,
      "auxiliary_loss_mlp": 0.01050024,
      "balance_loss_clip": 1.04081082,
      "balance_loss_mlp": 1.03217793,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 2.878226226781863,
      "language_loss": 0.77914679,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.80235505,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.6943821907043457
    },
    {
      "auxiliary_loss_clip": 0.01255078,
      "auxiliary_loss_mlp": 0.01055814,
      "balance_loss_clip": 1.03419507,
      "balance_loss_mlp": 1.03758693,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 2.0525650412979908,
      "language_loss": 0.85644323,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.87955213,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.7420601844787598
    },
    {
      "auxiliary_loss_clip": 0.0125992,
      "auxiliary_loss_mlp": 0.0104795,
      "balance_loss_clip": 0.95574617,
      "balance_loss_mlp": 1.03016448,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 1.8810621697786643,
      "language_loss": 0.68381286,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.70689154,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.7592196464538574
    },
    {
      "auxiliary_loss_clip": 0.01274107,
      "auxiliary_loss_mlp": 0.01131573,
      "balance_loss_clip": 1.03554249,
      "balance_loss_mlp": 0.0,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 1.9075521811176634,
      "language_loss": 0.73490655,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.75896335,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 2.6828911304473877
    },
    {
      "auxiliary_loss_clip": 0.01272734,
      "auxiliary_loss_mlp": 0.0104737,
      "balance_loss_clip": 1.0754329,
      "balance_loss_mlp": 1.02977443,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 2.137473529150511,
      "language_loss": 1.0027808,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.0259819,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 2.8478379249572754
    },
    {
      "auxiliary_loss_clip": 0.01269041,
      "auxiliary_loss_mlp": 0.01059185,
      "balance_loss_clip": 1.03618681,
      "balance_loss_mlp": 1.04154229,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.7179950537886197,
      "language_loss": 0.93572509,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.95900738,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.764920234680176
    },
    {
      "auxiliary_loss_clip": 0.01249705,
      "auxiliary_loss_mlp": 0.01058607,
      "balance_loss_clip": 0.99346697,
      "balance_loss_mlp": 1.04116702,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 1.7899757320092808,
      "language_loss": 0.67476285,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.69784594,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.7873117923736572
    },
    {
      "auxiliary_loss_clip": 0.01244419,
      "auxiliary_loss_mlp": 0.01050093,
      "balance_loss_clip": 0.91435623,
      "balance_loss_mlp": 1.03144908,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 8.462730363452577,
      "language_loss": 0.80982685,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.8327719,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.8430240154266357
    },
    {
      "auxiliary_loss_clip": 0.01259244,
      "auxiliary_loss_mlp": 0.01047043,
      "balance_loss_clip": 1.03434455,
      "balance_loss_mlp": 1.02956653,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 1.8187340095919184,
      "language_loss": 0.84360182,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.86666471,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 2.842942476272583
    },
    {
      "auxiliary_loss_clip": 0.01260985,
      "auxiliary_loss_mlp": 0.0113156,
      "balance_loss_clip": 0.99720216,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 1.9480125110161914,
      "language_loss": 0.82694197,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.85086739,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 2.771627187728882
    },
    {
      "auxiliary_loss_clip": 0.01178115,
      "auxiliary_loss_mlp": 0.01011035,
      "balance_loss_clip": 1.02006006,
      "balance_loss_mlp": 0.99873239,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7931072759708363,
      "language_loss": 0.6264416,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64833307,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 3.2282121181488037
    },
    {
      "auxiliary_loss_clip": 0.01255088,
      "auxiliary_loss_mlp": 0.010536,
      "balance_loss_clip": 0.99550986,
      "balance_loss_mlp": 1.03587413,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 2.6537204720720275,
      "language_loss": 0.86163008,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.88471699,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 2.7254421710968018
    },
    {
      "auxiliary_loss_clip": 0.0123322,
      "auxiliary_loss_mlp": 0.01045054,
      "balance_loss_clip": 0.94651741,
      "balance_loss_mlp": 1.02696955,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.4577131768341145,
      "language_loss": 0.81929857,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.84208131,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 3.570585012435913
    },
    {
      "auxiliary_loss_clip": 0.01226166,
      "auxiliary_loss_mlp": 0.01052439,
      "balance_loss_clip": 0.86735159,
      "balance_loss_mlp": 1.03461766,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 1.7619933242426622,
      "language_loss": 0.84457922,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.86736524,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 3.858222007751465
    },
    {
      "auxiliary_loss_clip": 0.01272758,
      "auxiliary_loss_mlp": 0.01051997,
      "balance_loss_clip": 1.03529644,
      "balance_loss_mlp": 1.03384185,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 2.3267299474648784,
      "language_loss": 0.87455869,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.89780617,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 3.6939754486083984
    },
    {
      "auxiliary_loss_clip": 0.01264914,
      "auxiliary_loss_mlp": 0.01052825,
      "balance_loss_clip": 0.99761355,
      "balance_loss_mlp": 1.03630304,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.6868547852663371,
      "language_loss": 0.86933935,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89251679,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 2.7321505546569824
    },
    {
      "auxiliary_loss_clip": 0.01269888,
      "auxiliary_loss_mlp": 0.01056062,
      "balance_loss_clip": 1.07306051,
      "balance_loss_mlp": 1.03857374,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.142800266105234,
      "language_loss": 0.95862693,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98188645,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 3.6180832386016846
    },
    {
      "auxiliary_loss_clip": 0.01262971,
      "auxiliary_loss_mlp": 0.01131286,
      "balance_loss_clip": 1.03299487,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 1.9780981944196234,
      "language_loss": 0.88273835,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.906681,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.6358699798583984
    },
    {
      "auxiliary_loss_clip": 0.01272167,
      "auxiliary_loss_mlp": 0.01055817,
      "balance_loss_clip": 1.07812285,
      "balance_loss_mlp": 1.03880632,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.6922984025782892,
      "language_loss": 0.96398252,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98726243,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 2.702413320541382
    },
    {
      "auxiliary_loss_clip": 0.01270376,
      "auxiliary_loss_mlp": 0.01131459,
      "balance_loss_clip": 1.07331204,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 2.4074966766514914,
      "language_loss": 0.81538361,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83940196,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.668992280960083
    },
    {
      "auxiliary_loss_clip": 0.0126992,
      "auxiliary_loss_mlp": 0.01056613,
      "balance_loss_clip": 1.07382703,
      "balance_loss_mlp": 1.03957796,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.366699679469943,
      "language_loss": 0.8740108,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.89727616,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.6983084678649902
    },
    {
      "auxiliary_loss_clip": 0.01263481,
      "auxiliary_loss_mlp": 0.01048053,
      "balance_loss_clip": 0.99691492,
      "balance_loss_mlp": 1.030792,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 2.7403230835779913,
      "language_loss": 0.77547938,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.79859471,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.734607458114624
    },
    {
      "auxiliary_loss_clip": 0.01260861,
      "auxiliary_loss_mlp": 0.0104886,
      "balance_loss_clip": 1.03502643,
      "balance_loss_mlp": 1.03264761,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.505323681994851,
      "language_loss": 0.70144886,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72454607,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.726377248764038
    },
    {
      "auxiliary_loss_clip": 0.01265487,
      "auxiliary_loss_mlp": 0.01048427,
      "balance_loss_clip": 0.91938657,
      "balance_loss_mlp": 1.02940083,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 2.3848443437097124,
      "language_loss": 0.8508817,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87402081,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.8457789421081543
    },
    {
      "auxiliary_loss_clip": 0.0126659,
      "auxiliary_loss_mlp": 0.01049358,
      "balance_loss_clip": 1.03388333,
      "balance_loss_mlp": 1.03167903,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 1.9316546663577134,
      "language_loss": 0.81191933,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83507884,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.697650194168091
    },
    {
      "auxiliary_loss_clip": 0.01259662,
      "auxiliary_loss_mlp": 0.01042218,
      "balance_loss_clip": 0.95341563,
      "balance_loss_mlp": 1.02575552,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 1.6594671984665834,
      "language_loss": 0.89952946,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92254823,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.905773878097534
    },
    {
      "auxiliary_loss_clip": 0.01265304,
      "auxiliary_loss_mlp": 0.01048648,
      "balance_loss_clip": 0.99145412,
      "balance_loss_mlp": 1.03116047,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 2.287109325708129,
      "language_loss": 0.77875942,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80189896,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.809784412384033
    },
    {
      "auxiliary_loss_clip": 0.01264415,
      "auxiliary_loss_mlp": 0.01046185,
      "balance_loss_clip": 0.99189031,
      "balance_loss_mlp": 1.02950764,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.880474285411748,
      "language_loss": 0.84847403,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.87158,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.7152013778686523
    },
    {
      "auxiliary_loss_clip": 0.01257155,
      "auxiliary_loss_mlp": 0.01131343,
      "balance_loss_clip": 1.02997255,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 2.9221862795791163,
      "language_loss": 0.85620201,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.88008696,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.7309224605560303
    },
    {
      "auxiliary_loss_clip": 0.0126447,
      "auxiliary_loss_mlp": 0.0106157,
      "balance_loss_clip": 0.99755502,
      "balance_loss_mlp": 1.04392719,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 1.8264049203441828,
      "language_loss": 0.80925286,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.83251321,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.706329107284546
    },
    {
      "auxiliary_loss_clip": 0.01264126,
      "auxiliary_loss_mlp": 0.01131202,
      "balance_loss_clip": 0.91782433,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 1.8264642129946218,
      "language_loss": 0.91979933,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.94375265,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 2.8734381198883057
    },
    {
      "auxiliary_loss_clip": 0.01261417,
      "auxiliary_loss_mlp": 0.01058415,
      "balance_loss_clip": 0.99570954,
      "balance_loss_mlp": 1.04102266,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.2798453745713045,
      "language_loss": 0.75866705,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78186536,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.841574192047119
    },
    {
      "auxiliary_loss_clip": 0.01267079,
      "auxiliary_loss_mlp": 0.01131275,
      "balance_loss_clip": 1.03020537,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 2.4584654563601287,
      "language_loss": 0.85775745,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.88174105,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.71313738822937
    },
    {
      "auxiliary_loss_clip": 0.01273007,
      "auxiliary_loss_mlp": 0.01045682,
      "balance_loss_clip": 1.07652199,
      "balance_loss_mlp": 1.02818227,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 1.7325489639606948,
      "language_loss": 0.84114468,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.8643316,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.6765706539154053
    },
    {
      "auxiliary_loss_clip": 0.01258548,
      "auxiliary_loss_mlp": 0.010573,
      "balance_loss_clip": 0.99689686,
      "balance_loss_mlp": 1.04015815,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.875327970492868,
      "language_loss": 0.86862254,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.89178097,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 2.835529327392578
    },
    {
      "auxiliary_loss_clip": 0.01272368,
      "auxiliary_loss_mlp": 0.01057655,
      "balance_loss_clip": 0.99615455,
      "balance_loss_mlp": 1.03964269,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 2.0317337344153543,
      "language_loss": 0.80219722,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82549751,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.7933669090270996
    },
    {
      "auxiliary_loss_clip": 0.01266544,
      "auxiliary_loss_mlp": 0.010579,
      "balance_loss_clip": 1.03619599,
      "balance_loss_mlp": 1.04017365,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.56014414923389,
      "language_loss": 0.72602677,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.74927121,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 2.7203080654144287
    },
    {
      "auxiliary_loss_clip": 0.01265714,
      "auxiliary_loss_mlp": 0.01048883,
      "balance_loss_clip": 1.00008893,
      "balance_loss_mlp": 1.03122783,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 3.0464334189789666,
      "language_loss": 0.75259328,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77573919,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 2.7615675926208496
    },
    {
      "auxiliary_loss_clip": 0.01272888,
      "auxiliary_loss_mlp": 0.01054405,
      "balance_loss_clip": 1.07698417,
      "balance_loss_mlp": 1.03703666,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 2.0228206250802683,
      "language_loss": 0.84434915,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.86762208,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 3.5412650108337402
    },
    {
      "auxiliary_loss_clip": 0.01250386,
      "auxiliary_loss_mlp": 0.01047221,
      "balance_loss_clip": 1.02955675,
      "balance_loss_mlp": 1.02964973,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 6.458787297329697,
      "language_loss": 0.84732103,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.87029707,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 3.6850905418395996
    },
    {
      "auxiliary_loss_clip": 0.0126237,
      "auxiliary_loss_mlp": 0.01060286,
      "balance_loss_clip": 0.9968276,
      "balance_loss_mlp": 1.04335845,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 2.407260342512204,
      "language_loss": 0.91675419,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.93998075,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 3.541778802871704
    },
    {
      "auxiliary_loss_clip": 0.01239818,
      "auxiliary_loss_mlp": 0.01130799,
      "balance_loss_clip": 0.95325315,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 1.950923122978381,
      "language_loss": 0.83203578,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.85574198,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 2.7219009399414062
    },
    {
      "auxiliary_loss_clip": 0.01254871,
      "auxiliary_loss_mlp": 0.01052115,
      "balance_loss_clip": 0.95301962,
      "balance_loss_mlp": 1.03436518,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 3.8587973956020867,
      "language_loss": 0.79942393,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.82249385,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 2.7551558017730713
    },
    {
      "auxiliary_loss_clip": 0.01268,
      "auxiliary_loss_mlp": 0.01047166,
      "balance_loss_clip": 1.0735234,
      "balance_loss_mlp": 1.03027439,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 2.2257169214914123,
      "language_loss": 0.73943722,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.76258886,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 3.8816018104553223
    },
    {
      "auxiliary_loss_clip": 0.01265592,
      "auxiliary_loss_mlp": 0.01047162,
      "balance_loss_clip": 0.95834744,
      "balance_loss_mlp": 1.02897084,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.481143250423937,
      "language_loss": 0.91784495,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.94097251,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 2.696971893310547
    },
    {
      "auxiliary_loss_clip": 0.01264199,
      "auxiliary_loss_mlp": 0.01046761,
      "balance_loss_clip": 0.99938542,
      "balance_loss_mlp": 1.02930832,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 3.978414885401318,
      "language_loss": 0.832421,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85553062,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.7215795516967773
    },
    {
      "auxiliary_loss_clip": 0.01267353,
      "auxiliary_loss_mlp": 0.0105279,
      "balance_loss_clip": 1.03814042,
      "balance_loss_mlp": 1.03482485,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 2.3673648293914966,
      "language_loss": 0.82443732,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84763879,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.658116340637207
    },
    {
      "auxiliary_loss_clip": 0.01253978,
      "auxiliary_loss_mlp": 0.01061648,
      "balance_loss_clip": 0.99405348,
      "balance_loss_mlp": 1.04525661,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 3.351288101714357,
      "language_loss": 0.83073002,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.85388625,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.733816146850586
    },
    {
      "auxiliary_loss_clip": 0.01254657,
      "auxiliary_loss_mlp": 0.01130928,
      "balance_loss_clip": 0.95453244,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.2797794868155226,
      "language_loss": 0.8100028,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.83385861,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.75335955619812
    },
    {
      "auxiliary_loss_clip": 0.01259409,
      "auxiliary_loss_mlp": 0.01054789,
      "balance_loss_clip": 0.92461336,
      "balance_loss_mlp": 1.03827894,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 1.8651833630096328,
      "language_loss": 0.87392437,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.89706635,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.803443670272827
    },
    {
      "auxiliary_loss_clip": 0.01263143,
      "auxiliary_loss_mlp": 0.0104916,
      "balance_loss_clip": 1.03258801,
      "balance_loss_mlp": 1.03126669,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 1.8682135167914917,
      "language_loss": 0.82286954,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.84599257,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.685316324234009
    },
    {
      "auxiliary_loss_clip": 0.01268852,
      "auxiliary_loss_mlp": 0.01048608,
      "balance_loss_clip": 1.07245779,
      "balance_loss_mlp": 1.03153694,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 2.1737142610115305,
      "language_loss": 0.74555671,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.76873136,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.758843183517456
    },
    {
      "auxiliary_loss_clip": 0.01272053,
      "auxiliary_loss_mlp": 0.01131243,
      "balance_loss_clip": 1.03390992,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 2.8766857717520065,
      "language_loss": 0.62294555,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.6469785,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.764472484588623
    },
    {
      "auxiliary_loss_clip": 0.01166556,
      "auxiliary_loss_mlp": 0.01009941,
      "balance_loss_clip": 0.90113026,
      "balance_loss_mlp": 0.99763834,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0152263228347107,
      "language_loss": 0.59469879,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61646378,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.3887147903442383
    },
    {
      "auxiliary_loss_clip": 0.01267451,
      "auxiliary_loss_mlp": 0.01047724,
      "balance_loss_clip": 1.07474661,
      "balance_loss_mlp": 1.03058147,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 3.313455586100102,
      "language_loss": 0.84075844,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.8639102,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.7318623065948486
    },
    {
      "auxiliary_loss_clip": 0.01267225,
      "auxiliary_loss_mlp": 0.01056561,
      "balance_loss_clip": 0.99670416,
      "balance_loss_mlp": 1.03904891,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.3485706081949784,
      "language_loss": 0.85452658,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.8777644,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.7952635288238525
    },
    {
      "auxiliary_loss_clip": 0.01219276,
      "auxiliary_loss_mlp": 0.01047246,
      "balance_loss_clip": 0.78794193,
      "balance_loss_mlp": 1.02991283,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.3592535410253164,
      "language_loss": 0.94493878,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96760398,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 3.3480186462402344
    },
    {
      "auxiliary_loss_clip": 0.01182301,
      "auxiliary_loss_mlp": 0.01014262,
      "balance_loss_clip": 1.06413531,
      "balance_loss_mlp": 1.00215006,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.8060934447526081,
      "language_loss": 0.55501735,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57698298,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.3123068809509277
    },
    {
      "auxiliary_loss_clip": 0.01248165,
      "auxiliary_loss_mlp": 0.0105356,
      "balance_loss_clip": 0.98933601,
      "balance_loss_mlp": 1.03645349,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 2.29602305102977,
      "language_loss": 0.81505013,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.83806741,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.9253787994384766
    },
    {
      "auxiliary_loss_clip": 0.01268695,
      "auxiliary_loss_mlp": 0.01044501,
      "balance_loss_clip": 1.07400143,
      "balance_loss_mlp": 1.02745438,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.5347613830529474,
      "language_loss": 0.81574464,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.83887661,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.786910057067871
    },
    {
      "auxiliary_loss_clip": 0.01260342,
      "auxiliary_loss_mlp": 0.0113105,
      "balance_loss_clip": 0.99614602,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 1.9516042374436828,
      "language_loss": 0.78188944,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.8058033,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.794328451156616
    },
    {
      "auxiliary_loss_clip": 0.01265518,
      "auxiliary_loss_mlp": 0.01055393,
      "balance_loss_clip": 1.03690326,
      "balance_loss_mlp": 1.03839421,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 1.998207212443615,
      "language_loss": 0.80953765,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.8327468,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 2.7559540271759033
    },
    {
      "auxiliary_loss_clip": 0.01255998,
      "auxiliary_loss_mlp": 0.01056145,
      "balance_loss_clip": 0.95536083,
      "balance_loss_mlp": 1.0384779,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 1.78830442130868,
      "language_loss": 0.77980059,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80292201,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 2.8905818462371826
    },
    {
      "auxiliary_loss_clip": 0.01262543,
      "auxiliary_loss_mlp": 0.01050627,
      "balance_loss_clip": 0.99878752,
      "balance_loss_mlp": 1.03279293,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.267486350759504,
      "language_loss": 0.91798013,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.94111186,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 2.686021566390991
    },
    {
      "auxiliary_loss_clip": 0.01257574,
      "auxiliary_loss_mlp": 0.01054085,
      "balance_loss_clip": 0.91667604,
      "balance_loss_mlp": 1.03738427,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.6574731412357806,
      "language_loss": 0.80296147,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82607806,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 3.808079719543457
    },
    {
      "auxiliary_loss_clip": 0.01267147,
      "auxiliary_loss_mlp": 0.01058279,
      "balance_loss_clip": 0.99483073,
      "balance_loss_mlp": 1.04110146,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 1.9869731582821268,
      "language_loss": 0.82506609,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.84832036,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 3.8824453353881836
    },
    {
      "auxiliary_loss_clip": 0.01266897,
      "auxiliary_loss_mlp": 0.01062398,
      "balance_loss_clip": 1.03720355,
      "balance_loss_mlp": 1.04431367,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 2.0593238382908314,
      "language_loss": 0.86689651,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89018947,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 3.6355888843536377
    },
    {
      "auxiliary_loss_clip": 0.01262119,
      "auxiliary_loss_mlp": 0.01060661,
      "balance_loss_clip": 0.99724424,
      "balance_loss_mlp": 1.0431726,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 2.8193279930915995,
      "language_loss": 0.84256756,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86579537,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 2.774378776550293
    },
    {
      "auxiliary_loss_clip": 0.01251939,
      "auxiliary_loss_mlp": 0.01056292,
      "balance_loss_clip": 1.02951968,
      "balance_loss_mlp": 1.03907835,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 3.294297628584271,
      "language_loss": 0.84387708,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86695933,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 3.7569937705993652
    },
    {
      "auxiliary_loss_clip": 0.01262283,
      "auxiliary_loss_mlp": 0.01053496,
      "balance_loss_clip": 0.96025538,
      "balance_loss_mlp": 1.0357697,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.157973361043339,
      "language_loss": 0.7428776,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76603544,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.701493501663208
    },
    {
      "auxiliary_loss_clip": 0.01260719,
      "auxiliary_loss_mlp": 0.01061819,
      "balance_loss_clip": 0.9579196,
      "balance_loss_mlp": 1.04437923,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.499839155186761,
      "language_loss": 0.85546654,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87869191,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.749748468399048
    },
    {
      "auxiliary_loss_clip": 0.01270636,
      "auxiliary_loss_mlp": 0.01055728,
      "balance_loss_clip": 1.03784823,
      "balance_loss_mlp": 1.03943193,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 2.2510507629638683,
      "language_loss": 0.91128057,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93454427,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.668240547180176
    },
    {
      "auxiliary_loss_clip": 0.01264907,
      "auxiliary_loss_mlp": 0.01052734,
      "balance_loss_clip": 0.95605993,
      "balance_loss_mlp": 1.03479266,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 1.8944602506948742,
      "language_loss": 0.72104329,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74421966,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.792232036590576
    },
    {
      "auxiliary_loss_clip": 0.01271447,
      "auxiliary_loss_mlp": 0.01052263,
      "balance_loss_clip": 1.07407629,
      "balance_loss_mlp": 1.03509688,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 2.278660349868121,
      "language_loss": 0.85939479,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.8826319,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.6557323932647705
    },
    {
      "auxiliary_loss_clip": 0.01270414,
      "auxiliary_loss_mlp": 0.01050847,
      "balance_loss_clip": 1.07271814,
      "balance_loss_mlp": 1.03341866,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 2.3589969703452223,
      "language_loss": 0.8861087,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.90932131,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.647742986679077
    },
    {
      "auxiliary_loss_clip": 0.0116378,
      "auxiliary_loss_mlp": 0.01014534,
      "balance_loss_clip": 0.97270346,
      "balance_loss_mlp": 1.00285196,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9885715547406374,
      "language_loss": 0.63583833,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.6576215,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.4845471382141113
    },
    {
      "auxiliary_loss_clip": 0.0126958,
      "auxiliary_loss_mlp": 0.010531,
      "balance_loss_clip": 1.0355494,
      "balance_loss_mlp": 1.03441954,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 3.1929721463498,
      "language_loss": 0.93814051,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.96136725,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.732806444168091
    },
    {
      "auxiliary_loss_clip": 0.01248737,
      "auxiliary_loss_mlp": 0.01055197,
      "balance_loss_clip": 0.95560503,
      "balance_loss_mlp": 1.0389483,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 1.9182799630556415,
      "language_loss": 0.84933209,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87237138,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.796555757522583
    },
    {
      "auxiliary_loss_clip": 0.01249253,
      "auxiliary_loss_mlp": 0.01052121,
      "balance_loss_clip": 1.03139365,
      "balance_loss_mlp": 1.03422737,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 1.922066214425766,
      "language_loss": 0.91645789,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.93947154,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.677762269973755
    },
    {
      "auxiliary_loss_clip": 0.01257777,
      "auxiliary_loss_mlp": 0.01048214,
      "balance_loss_clip": 0.95489997,
      "balance_loss_mlp": 1.03120244,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 2.763420360782473,
      "language_loss": 0.71149242,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73455232,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.8534724712371826
    },
    {
      "auxiliary_loss_clip": 0.01242819,
      "auxiliary_loss_mlp": 0.01056681,
      "balance_loss_clip": 0.95320469,
      "balance_loss_mlp": 1.04061198,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.6731048590838131,
      "language_loss": 0.91068667,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93368167,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.802180528640747
    },
    {
      "auxiliary_loss_clip": 0.01238612,
      "auxiliary_loss_mlp": 0.01131582,
      "balance_loss_clip": 0.87499779,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 2.3056998122880588,
      "language_loss": 1.00500047,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.0287025,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.8276190757751465
    },
    {
      "auxiliary_loss_clip": 0.01262182,
      "auxiliary_loss_mlp": 0.01051278,
      "balance_loss_clip": 0.99674249,
      "balance_loss_mlp": 1.03357506,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 3.1769302169235267,
      "language_loss": 0.74356806,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.76670265,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 2.714745044708252
    },
    {
      "auxiliary_loss_clip": 0.01253007,
      "auxiliary_loss_mlp": 0.01047119,
      "balance_loss_clip": 0.99479383,
      "balance_loss_mlp": 1.02910638,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.3544839308202286,
      "language_loss": 0.88946086,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91246212,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 2.8031370639801025
    },
    {
      "auxiliary_loss_clip": 0.01268221,
      "auxiliary_loss_mlp": 0.01131315,
      "balance_loss_clip": 1.07443023,
      "balance_loss_mlp": 0.0,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 3.3047996118561467,
      "language_loss": 0.91190428,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.93589962,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.6382367610931396
    },
    {
      "auxiliary_loss_clip": 0.01249221,
      "auxiliary_loss_mlp": 0.01050337,
      "balance_loss_clip": 0.95800471,
      "balance_loss_mlp": 1.03244376,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.1912159696956137,
      "language_loss": 0.865026,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.88802147,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.7613940238952637
    },
    {
      "auxiliary_loss_clip": 0.01268741,
      "auxiliary_loss_mlp": 0.01056276,
      "balance_loss_clip": 0.99836993,
      "balance_loss_mlp": 1.038728,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 3.143432295850078,
      "language_loss": 0.89374965,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91699982,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 2.736199378967285
    },
    {
      "auxiliary_loss_clip": 0.01266442,
      "auxiliary_loss_mlp": 0.01052355,
      "balance_loss_clip": 0.95845062,
      "balance_loss_mlp": 1.03479505,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 2.1418344509267357,
      "language_loss": 0.78051943,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.80370736,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.7394373416900635
    },
    {
      "auxiliary_loss_clip": 0.01250657,
      "auxiliary_loss_mlp": 0.01052692,
      "balance_loss_clip": 0.9925102,
      "balance_loss_mlp": 1.03465545,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 2.2319833498910477,
      "language_loss": 0.85895205,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.88198555,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 2.7751450538635254
    },
    {
      "auxiliary_loss_clip": 0.01254835,
      "auxiliary_loss_mlp": 0.01051095,
      "balance_loss_clip": 0.99407816,
      "balance_loss_mlp": 1.03321385,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 2.1049414174915357,
      "language_loss": 0.8534956,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.87655491,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 2.812270402908325
    },
    {
      "auxiliary_loss_clip": 0.01250664,
      "auxiliary_loss_mlp": 0.01047919,
      "balance_loss_clip": 0.95370591,
      "balance_loss_mlp": 1.03085995,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 2.4454532020315023,
      "language_loss": 0.99066585,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01365173,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 4.746300220489502
    },
    {
      "auxiliary_loss_clip": 0.01262207,
      "auxiliary_loss_mlp": 0.01050162,
      "balance_loss_clip": 0.96116072,
      "balance_loss_mlp": 1.03331745,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 2.2676032586129757,
      "language_loss": 0.75662392,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.77974761,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.7866628170013428
    },
    {
      "auxiliary_loss_clip": 0.01258713,
      "auxiliary_loss_mlp": 0.01058535,
      "balance_loss_clip": 0.91909051,
      "balance_loss_mlp": 1.04191685,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 1.8351396907288444,
      "language_loss": 0.75082207,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77399457,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 3.6791882514953613
    },
    {
      "auxiliary_loss_clip": 0.01270111,
      "auxiliary_loss_mlp": 0.01051276,
      "balance_loss_clip": 1.07512069,
      "balance_loss_mlp": 1.03404999,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 2.248448878223978,
      "language_loss": 0.85312736,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87634122,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 2.6429648399353027
    },
    {
      "auxiliary_loss_clip": 0.01270704,
      "auxiliary_loss_mlp": 0.01053979,
      "balance_loss_clip": 0.99609208,
      "balance_loss_mlp": 1.03618097,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.5735857567301523,
      "language_loss": 0.80916798,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.83241487,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 3.689455032348633
    },
    {
      "auxiliary_loss_clip": 0.01272178,
      "auxiliary_loss_mlp": 0.01054746,
      "balance_loss_clip": 1.03691566,
      "balance_loss_mlp": 1.03637576,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.683099262052174,
      "language_loss": 0.91599256,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.93926185,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 2.6984541416168213
    },
    {
      "auxiliary_loss_clip": 0.01253588,
      "auxiliary_loss_mlp": 0.01057986,
      "balance_loss_clip": 0.99640441,
      "balance_loss_mlp": 1.03961575,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 3.5629605022930857,
      "language_loss": 0.81764245,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.8407582,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.7050678730010986
    },
    {
      "auxiliary_loss_clip": 0.01263229,
      "auxiliary_loss_mlp": 0.01044853,
      "balance_loss_clip": 0.95346463,
      "balance_loss_mlp": 1.0266614,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 1.8243532106309717,
      "language_loss": 0.7895937,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81267446,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.7759976387023926
    },
    {
      "auxiliary_loss_clip": 0.01266727,
      "auxiliary_loss_mlp": 0.01051087,
      "balance_loss_clip": 1.03695154,
      "balance_loss_mlp": 1.03423071,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.1428396286782174,
      "language_loss": 0.83414257,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.85732073,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.670766830444336
    },
    {
      "auxiliary_loss_clip": 0.01266102,
      "auxiliary_loss_mlp": 0.01051595,
      "balance_loss_clip": 1.03374624,
      "balance_loss_mlp": 1.03434563,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.136359246824773,
      "language_loss": 0.7849648,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.80814177,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.9524409770965576
    },
    {
      "auxiliary_loss_clip": 0.01254019,
      "auxiliary_loss_mlp": 0.0104636,
      "balance_loss_clip": 0.9903394,
      "balance_loss_mlp": 1.02986145,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 1.753008782585102,
      "language_loss": 0.85150945,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87451321,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.7924139499664307
    },
    {
      "auxiliary_loss_clip": 0.01268151,
      "auxiliary_loss_mlp": 0.01052666,
      "balance_loss_clip": 1.07358265,
      "balance_loss_mlp": 1.03510618,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.408130921861515,
      "language_loss": 0.91877818,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.94198638,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.7372565269470215
    },
    {
      "auxiliary_loss_clip": 0.01260201,
      "auxiliary_loss_mlp": 0.01049109,
      "balance_loss_clip": 0.95662248,
      "balance_loss_mlp": 1.03216922,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 3.707020720478274,
      "language_loss": 0.78684008,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.80993313,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.8038647174835205
    },
    {
      "auxiliary_loss_clip": 0.01265814,
      "auxiliary_loss_mlp": 0.01049129,
      "balance_loss_clip": 1.03348327,
      "balance_loss_mlp": 1.03234458,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 1.965130452177894,
      "language_loss": 0.92440784,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94755727,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.7989003658294678
    },
    {
      "auxiliary_loss_clip": 0.01247493,
      "auxiliary_loss_mlp": 0.01055646,
      "balance_loss_clip": 0.95575249,
      "balance_loss_mlp": 1.03771722,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 2.555068107985232,
      "language_loss": 0.73019779,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.75322914,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.76094126701355
    },
    {
      "auxiliary_loss_clip": 0.01265968,
      "auxiliary_loss_mlp": 0.01047182,
      "balance_loss_clip": 1.03558826,
      "balance_loss_mlp": 1.03020656,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 2.1232289533358633,
      "language_loss": 0.92148256,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.94461405,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.74515962600708
    },
    {
      "auxiliary_loss_clip": 0.01252947,
      "auxiliary_loss_mlp": 0.01060107,
      "balance_loss_clip": 0.95388544,
      "balance_loss_mlp": 1.04246426,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 1.8785856397870275,
      "language_loss": 0.87581372,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.89894432,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.8320789337158203
    },
    {
      "auxiliary_loss_clip": 0.01260601,
      "auxiliary_loss_mlp": 0.01052002,
      "balance_loss_clip": 0.95346022,
      "balance_loss_mlp": 1.03474033,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 2.6289587491493327,
      "language_loss": 0.85338145,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87650746,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.8157060146331787
    },
    {
      "auxiliary_loss_clip": 0.01258604,
      "auxiliary_loss_mlp": 0.01048422,
      "balance_loss_clip": 0.99125409,
      "balance_loss_mlp": 1.03130364,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 3.673717152773359,
      "language_loss": 0.68409705,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.70716733,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.9129364490509033
    },
    {
      "auxiliary_loss_clip": 0.01265182,
      "auxiliary_loss_mlp": 0.0104721,
      "balance_loss_clip": 0.99474716,
      "balance_loss_mlp": 1.0300914,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 1.8863369586182699,
      "language_loss": 0.81540763,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.83853155,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 2.7328097820281982
    },
    {
      "auxiliary_loss_clip": 0.01238195,
      "auxiliary_loss_mlp": 0.01055754,
      "balance_loss_clip": 0.94984907,
      "balance_loss_mlp": 1.03960133,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 2.124443629611172,
      "language_loss": 0.86343396,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.8863734,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.8593130111694336
    },
    {
      "auxiliary_loss_clip": 0.01259129,
      "auxiliary_loss_mlp": 0.01053102,
      "balance_loss_clip": 0.91651177,
      "balance_loss_mlp": 1.0365324,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 1.6522815795475891,
      "language_loss": 0.83230329,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.8554256,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.8147239685058594
    },
    {
      "auxiliary_loss_clip": 0.01272718,
      "auxiliary_loss_mlp": 0.01060648,
      "balance_loss_clip": 1.00341678,
      "balance_loss_mlp": 1.04391098,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 1.8188881272402397,
      "language_loss": 0.85603511,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.87936878,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 2.819133758544922
    },
    {
      "auxiliary_loss_clip": 0.01268893,
      "auxiliary_loss_mlp": 0.01049604,
      "balance_loss_clip": 1.03769922,
      "balance_loss_mlp": 1.031901,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 1.9948913114410196,
      "language_loss": 0.82390183,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.84708679,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.711665153503418
    },
    {
      "auxiliary_loss_clip": 0.01248644,
      "auxiliary_loss_mlp": 0.01054912,
      "balance_loss_clip": 1.03273928,
      "balance_loss_mlp": 1.03804398,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.690282196492058,
      "language_loss": 0.83032364,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.85335922,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.663419008255005
    },
    {
      "auxiliary_loss_clip": 0.01258979,
      "auxiliary_loss_mlp": 0.0104213,
      "balance_loss_clip": 0.99529707,
      "balance_loss_mlp": 1.02464175,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 2.1510145335987447,
      "language_loss": 0.86424339,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88725448,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 2.7453248500823975
    },
    {
      "auxiliary_loss_clip": 0.01264252,
      "auxiliary_loss_mlp": 0.0104933,
      "balance_loss_clip": 1.03501213,
      "balance_loss_mlp": 1.03210425,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 2.475560403442135,
      "language_loss": 0.68267107,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70580691,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 4.719552516937256
    },
    {
      "auxiliary_loss_clip": 0.01261981,
      "auxiliary_loss_mlp": 0.01056955,
      "balance_loss_clip": 1.03248668,
      "balance_loss_mlp": 1.03970551,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.2346730559379235,
      "language_loss": 0.88336307,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90655243,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 2.801924228668213
    },
    {
      "auxiliary_loss_clip": 0.01247332,
      "auxiliary_loss_mlp": 0.01047261,
      "balance_loss_clip": 1.0304811,
      "balance_loss_mlp": 1.03029728,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 1.9381002110534498,
      "language_loss": 0.87794816,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90089411,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 3.64528226852417
    },
    {
      "auxiliary_loss_clip": 0.0126729,
      "auxiliary_loss_mlp": 0.01051801,
      "balance_loss_clip": 1.07554817,
      "balance_loss_mlp": 1.03545773,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 1.6145922241432027,
      "language_loss": 0.77912319,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80231404,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 2.6639058589935303
    },
    {
      "auxiliary_loss_clip": 0.01259864,
      "auxiliary_loss_mlp": 0.01051577,
      "balance_loss_clip": 0.91905648,
      "balance_loss_mlp": 1.03526926,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 4.277408176208244,
      "language_loss": 0.87130606,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.8944205,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 2.8860011100769043
    },
    {
      "auxiliary_loss_clip": 0.01260267,
      "auxiliary_loss_mlp": 0.01045625,
      "balance_loss_clip": 1.03145087,
      "balance_loss_mlp": 1.02930498,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 1.9149212169768148,
      "language_loss": 0.84247881,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86553776,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 3.7589035034179688
    },
    {
      "auxiliary_loss_clip": 0.01255922,
      "auxiliary_loss_mlp": 0.01041897,
      "balance_loss_clip": 0.95647907,
      "balance_loss_mlp": 1.02539861,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.6748925502165584,
      "language_loss": 0.76647806,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78945625,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 2.799381971359253
    },
    {
      "auxiliary_loss_clip": 0.01256747,
      "auxiliary_loss_mlp": 0.01044702,
      "balance_loss_clip": 1.03285646,
      "balance_loss_mlp": 1.02761936,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 2.1056944406443736,
      "language_loss": 0.89363223,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91664672,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.7628540992736816
    },
    {
      "auxiliary_loss_clip": 0.01250669,
      "auxiliary_loss_mlp": 0.0105803,
      "balance_loss_clip": 0.91710323,
      "balance_loss_mlp": 1.04119742,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.2942310213939465,
      "language_loss": 0.82949448,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.85258138,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.782944917678833
    },
    {
      "auxiliary_loss_clip": 0.01240699,
      "auxiliary_loss_mlp": 0.01056505,
      "balance_loss_clip": 0.95353472,
      "balance_loss_mlp": 1.03913593,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 2.649241896050479,
      "language_loss": 0.84604919,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.86902118,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.7997283935546875
    },
    {
      "auxiliary_loss_clip": 0.0124945,
      "auxiliary_loss_mlp": 0.01040365,
      "balance_loss_clip": 0.98912424,
      "balance_loss_mlp": 1.0238781,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 2.0584435928045224,
      "language_loss": 0.93707961,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95997775,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.7377164363861084
    },
    {
      "auxiliary_loss_clip": 0.0124271,
      "auxiliary_loss_mlp": 0.01131048,
      "balance_loss_clip": 0.9506129,
      "balance_loss_mlp": 0.0,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 2.2399285145587857,
      "language_loss": 0.75508481,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.77882236,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.803431510925293
    },
    {
      "auxiliary_loss_clip": 0.01250569,
      "auxiliary_loss_mlp": 0.01054794,
      "balance_loss_clip": 0.91276205,
      "balance_loss_mlp": 1.03815281,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.864050349138455,
      "language_loss": 0.78167284,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80472648,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.7752628326416016
    },
    {
      "auxiliary_loss_clip": 0.01251931,
      "auxiliary_loss_mlp": 0.01130636,
      "balance_loss_clip": 0.99055046,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 2.218380056184846,
      "language_loss": 0.89506686,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.9188925,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.7660393714904785
    },
    {
      "auxiliary_loss_clip": 0.01258173,
      "auxiliary_loss_mlp": 0.0104915,
      "balance_loss_clip": 0.99863434,
      "balance_loss_mlp": 1.03317571,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 2.796467738578854,
      "language_loss": 0.64080739,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66388065,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.8832929134368896
    },
    {
      "auxiliary_loss_clip": 0.01257035,
      "auxiliary_loss_mlp": 0.01054451,
      "balance_loss_clip": 1.03406799,
      "balance_loss_mlp": 1.03864431,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.17452727789309,
      "language_loss": 0.78246021,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80557501,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.678382635116577
    },
    {
      "auxiliary_loss_clip": 0.01247427,
      "auxiliary_loss_mlp": 0.01048185,
      "balance_loss_clip": 0.9116807,
      "balance_loss_mlp": 1.03186512,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.184518251385324,
      "language_loss": 0.9058609,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.92881703,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.7694814205169678
    },
    {
      "auxiliary_loss_clip": 0.01252595,
      "auxiliary_loss_mlp": 0.01046281,
      "balance_loss_clip": 0.8369993,
      "balance_loss_mlp": 1.03021133,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 1.8804995145456673,
      "language_loss": 0.86317945,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.88616824,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 3.0398683547973633
    },
    {
      "auxiliary_loss_clip": 0.01258128,
      "auxiliary_loss_mlp": 0.01047153,
      "balance_loss_clip": 0.95496827,
      "balance_loss_mlp": 1.03052354,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 4.7623049401052215,
      "language_loss": 0.86490172,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88795453,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.930471897125244
    },
    {
      "auxiliary_loss_clip": 0.01259209,
      "auxiliary_loss_mlp": 0.01052046,
      "balance_loss_clip": 1.06808078,
      "balance_loss_mlp": 1.03666842,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.2484292466500158,
      "language_loss": 0.87605321,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.89916581,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.6944739818573
    },
    {
      "auxiliary_loss_clip": 0.01178596,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 0.98272991,
      "balance_loss_mlp": 1.01707578,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8213683794440432,
      "language_loss": 0.555861,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57792747,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 3.4703855514526367
    },
    {
      "auxiliary_loss_clip": 0.01236651,
      "auxiliary_loss_mlp": 0.01053141,
      "balance_loss_clip": 0.90901089,
      "balance_loss_mlp": 1.03667831,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 1.9621769293042917,
      "language_loss": 0.87674975,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.89964765,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.8550052642822266
    },
    {
      "auxiliary_loss_clip": 0.01251645,
      "auxiliary_loss_mlp": 0.01053759,
      "balance_loss_clip": 0.95769513,
      "balance_loss_mlp": 1.03734422,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 2.0200989983257953,
      "language_loss": 0.88442922,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.90748334,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.7616677284240723
    },
    {
      "auxiliary_loss_clip": 0.01256413,
      "auxiliary_loss_mlp": 0.01048691,
      "balance_loss_clip": 1.03245509,
      "balance_loss_mlp": 1.03314638,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 3.1328791580204265,
      "language_loss": 0.73852563,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76157665,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.7761011123657227
    },
    {
      "auxiliary_loss_clip": 0.01257698,
      "auxiliary_loss_mlp": 0.01051342,
      "balance_loss_clip": 1.03324533,
      "balance_loss_mlp": 1.03512931,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 2.3136301891923785,
      "language_loss": 0.89352143,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.91661185,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 2.734149217605591
    },
    {
      "auxiliary_loss_clip": 0.01167893,
      "auxiliary_loss_mlp": 0.01013184,
      "balance_loss_clip": 1.01977015,
      "balance_loss_mlp": 1.00231206,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8720489535245326,
      "language_loss": 0.62656844,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64837921,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 3.132209539413452
    },
    {
      "auxiliary_loss_clip": 0.01237055,
      "auxiliary_loss_mlp": 0.01055536,
      "balance_loss_clip": 0.95107353,
      "balance_loss_mlp": 1.03928804,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.193494924544307,
      "language_loss": 0.88720095,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.91012686,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 3.7327733039855957
    },
    {
      "auxiliary_loss_clip": 0.01254604,
      "auxiliary_loss_mlp": 0.01130394,
      "balance_loss_clip": 1.03198981,
      "balance_loss_mlp": 0.0,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 2.501675932673333,
      "language_loss": 0.82134503,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.84519494,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 2.832578659057617
    },
    {
      "auxiliary_loss_clip": 0.01165316,
      "auxiliary_loss_mlp": 0.01011623,
      "balance_loss_clip": 0.93863928,
      "balance_loss_mlp": 1.00084686,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8569311330217394,
      "language_loss": 0.61423385,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63600326,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 4.14725136756897
    },
    {
      "auxiliary_loss_clip": 0.01258594,
      "auxiliary_loss_mlp": 0.01044407,
      "balance_loss_clip": 1.02990532,
      "balance_loss_mlp": 1.02852821,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 1.7473004498216245,
      "language_loss": 0.82748711,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.85051709,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 2.85221529006958
    },
    {
      "auxiliary_loss_clip": 0.01260509,
      "auxiliary_loss_mlp": 0.01048528,
      "balance_loss_clip": 1.02996671,
      "balance_loss_mlp": 1.03298306,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 1.7191263382084632,
      "language_loss": 0.88394725,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.90703762,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.72139835357666
    },
    {
      "auxiliary_loss_clip": 0.01252959,
      "auxiliary_loss_mlp": 0.01052534,
      "balance_loss_clip": 0.99371457,
      "balance_loss_mlp": 1.03671455,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 9.10804814964547,
      "language_loss": 0.83161706,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85467196,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 3.7617712020874023
    },
    {
      "auxiliary_loss_clip": 0.01256431,
      "auxiliary_loss_mlp": 0.01046807,
      "balance_loss_clip": 1.02901828,
      "balance_loss_mlp": 1.03111899,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 2.0461197330297565,
      "language_loss": 0.87067342,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.89370573,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.7341864109039307
    },
    {
      "auxiliary_loss_clip": 0.01251556,
      "auxiliary_loss_mlp": 0.01045376,
      "balance_loss_clip": 1.02782345,
      "balance_loss_mlp": 1.02917492,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 2.1862413579019337,
      "language_loss": 0.81204373,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83501303,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.660305976867676
    },
    {
      "auxiliary_loss_clip": 0.01260494,
      "auxiliary_loss_mlp": 0.01049084,
      "balance_loss_clip": 1.06890619,
      "balance_loss_mlp": 1.03240681,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 2.1988989956140546,
      "language_loss": 0.74538088,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76847661,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.6846816539764404
    },
    {
      "auxiliary_loss_clip": 0.01247039,
      "auxiliary_loss_mlp": 0.0104519,
      "balance_loss_clip": 0.98716027,
      "balance_loss_mlp": 1.02912092,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 2.105555601895299,
      "language_loss": 0.75646406,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.77938628,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.821315288543701
    },
    {
      "auxiliary_loss_clip": 0.0125674,
      "auxiliary_loss_mlp": 0.01052356,
      "balance_loss_clip": 0.99370694,
      "balance_loss_mlp": 1.03557169,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 2.639716447549593,
      "language_loss": 0.87179446,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89488542,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.7339906692504883
    },
    {
      "auxiliary_loss_clip": 0.01239846,
      "auxiliary_loss_mlp": 0.01044448,
      "balance_loss_clip": 0.98944414,
      "balance_loss_mlp": 1.02820015,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.1428095445214828,
      "language_loss": 0.76822686,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.79106975,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.7319061756134033
    },
    {
      "auxiliary_loss_clip": 0.01172089,
      "auxiliary_loss_mlp": 0.01009871,
      "balance_loss_clip": 1.01483548,
      "balance_loss_mlp": 0.99885559,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.9032321053118368,
      "language_loss": 0.6457926,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.6676122,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.296032190322876
    },
    {
      "auxiliary_loss_clip": 0.0125605,
      "auxiliary_loss_mlp": 0.01055572,
      "balance_loss_clip": 1.07121754,
      "balance_loss_mlp": 1.03945458,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 2.2173239316595663,
      "language_loss": 0.83961427,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.8627305,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.686323881149292
    },
    {
      "auxiliary_loss_clip": 0.01163583,
      "auxiliary_loss_mlp": 0.01009172,
      "balance_loss_clip": 1.01190019,
      "balance_loss_mlp": 0.99782318,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.8864229900990372,
      "language_loss": 0.64640856,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66813612,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.1993069648742676
    },
    {
      "auxiliary_loss_clip": 0.01253517,
      "auxiliary_loss_mlp": 0.01049803,
      "balance_loss_clip": 1.02845144,
      "balance_loss_mlp": 1.03416324,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 3.065368272820675,
      "language_loss": 0.84317172,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86620492,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.7644100189208984
    },
    {
      "auxiliary_loss_clip": 0.01260068,
      "auxiliary_loss_mlp": 0.01045437,
      "balance_loss_clip": 1.07234359,
      "balance_loss_mlp": 1.02912927,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 2.6716861837636454,
      "language_loss": 0.86810148,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.89115655,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.703094005584717
    },
    {
      "auxiliary_loss_clip": 0.0124877,
      "auxiliary_loss_mlp": 0.01050447,
      "balance_loss_clip": 0.95434439,
      "balance_loss_mlp": 1.03490257,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 1.9851352626678849,
      "language_loss": 0.80662352,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.82961571,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.7667667865753174
    },
    {
      "auxiliary_loss_clip": 0.0124718,
      "auxiliary_loss_mlp": 0.01048001,
      "balance_loss_clip": 0.95470476,
      "balance_loss_mlp": 1.03160977,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 1.7624336736134576,
      "language_loss": 0.82958102,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.85253286,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.7834839820861816
    },
    {
      "auxiliary_loss_clip": 0.01164101,
      "auxiliary_loss_mlp": 0.01012872,
      "balance_loss_clip": 0.93857431,
      "balance_loss_mlp": 1.00238168,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8444990103102271,
      "language_loss": 0.60042846,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62219822,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.361875295639038
    },
    {
      "auxiliary_loss_clip": 0.01248952,
      "auxiliary_loss_mlp": 0.01057242,
      "balance_loss_clip": 0.98930657,
      "balance_loss_mlp": 1.04167318,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.5707090547513323,
      "language_loss": 0.60984063,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63290262,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 2.8956737518310547
    },
    {
      "auxiliary_loss_clip": 0.01255002,
      "auxiliary_loss_mlp": 0.01049589,
      "balance_loss_clip": 1.03094745,
      "balance_loss_mlp": 1.0339011,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 2.001387031995149,
      "language_loss": 0.80590063,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.82894653,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 2.6863505840301514
    },
    {
      "auxiliary_loss_clip": 0.01258104,
      "auxiliary_loss_mlp": 0.01056538,
      "balance_loss_clip": 0.99461544,
      "balance_loss_mlp": 1.03933597,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 1.7377696426186904,
      "language_loss": 0.67874634,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.70189285,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.784518241882324
    },
    {
      "auxiliary_loss_clip": 0.01251706,
      "auxiliary_loss_mlp": 0.01053588,
      "balance_loss_clip": 0.99052262,
      "balance_loss_mlp": 1.03827,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 2.0793482598491795,
      "language_loss": 0.81600624,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83905923,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.7490389347076416
    },
    {
      "auxiliary_loss_clip": 0.01261877,
      "auxiliary_loss_mlp": 0.01059945,
      "balance_loss_clip": 1.03585672,
      "balance_loss_mlp": 1.0436368,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.6044303433621896,
      "language_loss": 0.66935521,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69257349,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.8413374423980713
    },
    {
      "auxiliary_loss_clip": 0.01234925,
      "auxiliary_loss_mlp": 0.01056127,
      "balance_loss_clip": 0.95288336,
      "balance_loss_mlp": 1.04113078,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 1.8143899042920868,
      "language_loss": 0.78017437,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80308491,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 2.7332496643066406
    },
    {
      "auxiliary_loss_clip": 0.01252576,
      "auxiliary_loss_mlp": 0.01046626,
      "balance_loss_clip": 0.99432433,
      "balance_loss_mlp": 1.03085434,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 12.403379284348157,
      "language_loss": 0.90733236,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.93032438,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 4.617533445358276
    },
    {
      "auxiliary_loss_clip": 0.01262266,
      "auxiliary_loss_mlp": 0.01055326,
      "balance_loss_clip": 1.07209837,
      "balance_loss_mlp": 1.03981662,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 2.294954503751395,
      "language_loss": 0.89120007,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91437602,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.680729627609253
    },
    {
      "auxiliary_loss_clip": 0.01253704,
      "auxiliary_loss_mlp": 0.01043724,
      "balance_loss_clip": 1.03320074,
      "balance_loss_mlp": 1.0286206,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 2.0526466434111086,
      "language_loss": 0.8307057,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85368001,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 3.7206218242645264
    },
    {
      "auxiliary_loss_clip": 0.01165083,
      "auxiliary_loss_mlp": 0.01016363,
      "balance_loss_clip": 1.0519737,
      "balance_loss_mlp": 1.00534773,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7424922231958306,
      "language_loss": 0.61246049,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63427484,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 3.13033390045166
    },
    {
      "auxiliary_loss_clip": 0.01253737,
      "auxiliary_loss_mlp": 0.0105088,
      "balance_loss_clip": 0.9551934,
      "balance_loss_mlp": 1.03420234,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 1.934376535942244,
      "language_loss": 0.84994602,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87299216,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 3.7921817302703857
    },
    {
      "auxiliary_loss_clip": 0.01253123,
      "auxiliary_loss_mlp": 0.01048953,
      "balance_loss_clip": 1.031551,
      "balance_loss_mlp": 1.03369439,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 2.1252243613920427,
      "language_loss": 0.67949498,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70251578,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 2.6708061695098877
    },
    {
      "auxiliary_loss_clip": 0.01252043,
      "auxiliary_loss_mlp": 0.01043129,
      "balance_loss_clip": 0.95693398,
      "balance_loss_mlp": 1.0268445,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 2.5168721785854453,
      "language_loss": 0.82002354,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84297526,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.8737611770629883
    },
    {
      "auxiliary_loss_clip": 0.0125484,
      "auxiliary_loss_mlp": 0.0104989,
      "balance_loss_clip": 1.03231883,
      "balance_loss_mlp": 1.03392804,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.3435770537841085,
      "language_loss": 0.82610607,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.8491534,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.7245750427246094
    },
    {
      "auxiliary_loss_clip": 0.01260145,
      "auxiliary_loss_mlp": 0.01049087,
      "balance_loss_clip": 0.99398768,
      "balance_loss_mlp": 1.03409064,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 1.947453051027106,
      "language_loss": 0.75796068,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.78105295,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.8391215801239014
    },
    {
      "auxiliary_loss_clip": 0.01250227,
      "auxiliary_loss_mlp": 0.01042804,
      "balance_loss_clip": 0.98898852,
      "balance_loss_mlp": 1.0271039,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 2.0488095614387896,
      "language_loss": 0.7766813,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.79961157,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.7637486457824707
    },
    {
      "auxiliary_loss_clip": 0.01227034,
      "auxiliary_loss_mlp": 0.01039802,
      "balance_loss_clip": 0.90646458,
      "balance_loss_mlp": 1.02431631,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 2.3521157493117335,
      "language_loss": 0.72632384,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.7489922,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.7039477825164795
    },
    {
      "auxiliary_loss_clip": 0.01255859,
      "auxiliary_loss_mlp": 0.01042572,
      "balance_loss_clip": 1.06900299,
      "balance_loss_mlp": 1.02732515,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 1.6832484488915613,
      "language_loss": 0.85717642,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.88016069,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.670947551727295
    },
    {
      "auxiliary_loss_clip": 0.012397,
      "auxiliary_loss_mlp": 0.01046434,
      "balance_loss_clip": 1.02714813,
      "balance_loss_mlp": 1.03103197,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 2.9335736259577523,
      "language_loss": 0.75132698,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77418834,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.687638759613037
    },
    {
      "auxiliary_loss_clip": 0.01247469,
      "auxiliary_loss_mlp": 0.01040492,
      "balance_loss_clip": 0.99055314,
      "balance_loss_mlp": 1.02525723,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 1.6485575986170853,
      "language_loss": 0.78963619,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.8125158,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.7657337188720703
    },
    {
      "auxiliary_loss_clip": 0.01250769,
      "auxiliary_loss_mlp": 0.01044837,
      "balance_loss_clip": 0.98990798,
      "balance_loss_mlp": 1.0296979,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.4487766925061036,
      "language_loss": 0.79794109,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.8208971,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.7249398231506348
    },
    {
      "auxiliary_loss_clip": 0.01244645,
      "auxiliary_loss_mlp": 0.0104908,
      "balance_loss_clip": 1.02886939,
      "balance_loss_mlp": 1.03290367,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 3.772029654855619,
      "language_loss": 0.77075875,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.79369605,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.6789979934692383
    },
    {
      "auxiliary_loss_clip": 0.01254565,
      "auxiliary_loss_mlp": 0.01054206,
      "balance_loss_clip": 1.03056777,
      "balance_loss_mlp": 1.0396384,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 3.9685902775088855,
      "language_loss": 0.72461772,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.74770546,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.777009963989258
    },
    {
      "auxiliary_loss_clip": 0.01249049,
      "auxiliary_loss_mlp": 0.01039851,
      "balance_loss_clip": 0.95159256,
      "balance_loss_mlp": 1.02509344,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 2.37672112965199,
      "language_loss": 0.88568258,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.9085716,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.753866195678711
    },
    {
      "auxiliary_loss_clip": 0.01248406,
      "auxiliary_loss_mlp": 0.01044908,
      "balance_loss_clip": 0.99055207,
      "balance_loss_mlp": 1.03090131,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 2.05274035983202,
      "language_loss": 0.72653234,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.74946547,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.6787283420562744
    },
    {
      "auxiliary_loss_clip": 0.01252049,
      "auxiliary_loss_mlp": 0.01043372,
      "balance_loss_clip": 0.95538735,
      "balance_loss_mlp": 1.02888787,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 1.859949585362936,
      "language_loss": 0.83401287,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.85696709,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.775991678237915
    },
    {
      "auxiliary_loss_clip": 0.01231825,
      "auxiliary_loss_mlp": 0.01041456,
      "balance_loss_clip": 0.98556995,
      "balance_loss_mlp": 1.0264715,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 2.517213594406481,
      "language_loss": 0.77272713,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.79545999,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.683511972427368
    },
    {
      "auxiliary_loss_clip": 0.0123147,
      "auxiliary_loss_mlp": 0.01045572,
      "balance_loss_clip": 0.98806393,
      "balance_loss_mlp": 1.03089786,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.191626288548549,
      "language_loss": 0.83177352,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85454392,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.6702938079833984
    },
    {
      "auxiliary_loss_clip": 0.01160383,
      "auxiliary_loss_mlp": 0.01011188,
      "balance_loss_clip": 1.00759768,
      "balance_loss_mlp": 1.00088811,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7386901054948797,
      "language_loss": 0.58821422,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60992992,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 3.394482374191284
    },
    {
      "auxiliary_loss_clip": 0.01227758,
      "auxiliary_loss_mlp": 0.01041894,
      "balance_loss_clip": 0.83180749,
      "balance_loss_mlp": 1.0262773,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 2.02180339484846,
      "language_loss": 0.88294053,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.90563703,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.800373077392578
    },
    {
      "auxiliary_loss_clip": 0.01235422,
      "auxiliary_loss_mlp": 0.01042443,
      "balance_loss_clip": 0.99048054,
      "balance_loss_mlp": 1.02694607,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 3.6949962946142563,
      "language_loss": 0.78748375,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.81026244,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 2.6895720958709717
    },
    {
      "auxiliary_loss_clip": 0.01235753,
      "auxiliary_loss_mlp": 0.01047446,
      "balance_loss_clip": 0.99203199,
      "balance_loss_mlp": 1.03156757,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 2.2023018346334933,
      "language_loss": 0.83459705,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85742909,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 2.69238018989563
    },
    {
      "auxiliary_loss_clip": 0.0125007,
      "auxiliary_loss_mlp": 0.01051868,
      "balance_loss_clip": 1.03170669,
      "balance_loss_mlp": 1.0366925,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 2.2890327468756575,
      "language_loss": 0.93460566,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.95762503,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 3.6273553371429443
    },
    {
      "auxiliary_loss_clip": 0.01256127,
      "auxiliary_loss_mlp": 0.01047519,
      "balance_loss_clip": 1.07170081,
      "balance_loss_mlp": 1.03240287,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 7.221685661537401,
      "language_loss": 0.78647238,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.8095088,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 2.7372355461120605
    },
    {
      "auxiliary_loss_clip": 0.01255992,
      "auxiliary_loss_mlp": 0.01045168,
      "balance_loss_clip": 1.03173816,
      "balance_loss_mlp": 1.03002834,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 2.2554481299106026,
      "language_loss": 0.83129728,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85430884,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 2.6144673824310303
    },
    {
      "auxiliary_loss_clip": 0.01248355,
      "auxiliary_loss_mlp": 0.01058806,
      "balance_loss_clip": 0.98790056,
      "balance_loss_mlp": 1.04398894,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 2.5113891006107525,
      "language_loss": 0.72697204,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.75004363,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 3.568223237991333
    },
    {
      "auxiliary_loss_clip": 0.01258682,
      "auxiliary_loss_mlp": 0.01046584,
      "balance_loss_clip": 1.0719707,
      "balance_loss_mlp": 1.0313015,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.096441235058588,
      "language_loss": 0.80125839,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82431108,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.7246439456939697
    },
    {
      "auxiliary_loss_clip": 0.01246036,
      "auxiliary_loss_mlp": 0.01045278,
      "balance_loss_clip": 0.99001729,
      "balance_loss_mlp": 1.03066325,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 2.0075321082273927,
      "language_loss": 0.89805746,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.92097062,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 3.6878504753112793
    },
    {
      "auxiliary_loss_clip": 0.01257114,
      "auxiliary_loss_mlp": 0.01046527,
      "balance_loss_clip": 0.9895367,
      "balance_loss_mlp": 1.0315547,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 2.0480395606931063,
      "language_loss": 0.84331459,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86635101,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.718367338180542
    },
    {
      "auxiliary_loss_clip": 0.01250986,
      "auxiliary_loss_mlp": 0.0105374,
      "balance_loss_clip": 0.99280322,
      "balance_loss_mlp": 1.03848076,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.5771998707865118,
      "language_loss": 0.89483976,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.91788703,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.7659237384796143
    },
    {
      "auxiliary_loss_clip": 0.01252269,
      "auxiliary_loss_mlp": 0.01129945,
      "balance_loss_clip": 1.03601539,
      "balance_loss_mlp": 0.0,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 2.780498432348566,
      "language_loss": 0.8223325,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84615469,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.6547601222991943
    },
    {
      "auxiliary_loss_clip": 0.01252482,
      "auxiliary_loss_mlp": 0.01055612,
      "balance_loss_clip": 0.99232745,
      "balance_loss_mlp": 1.03984046,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 3.398154640450128,
      "language_loss": 0.77567321,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79875416,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.650268793106079
    },
    {
      "auxiliary_loss_clip": 0.01251243,
      "auxiliary_loss_mlp": 0.01129825,
      "balance_loss_clip": 0.99363422,
      "balance_loss_mlp": 0.0,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 2.0775835532470373,
      "language_loss": 0.74730796,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.77111864,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.695493221282959
    },
    {
      "auxiliary_loss_clip": 0.01262383,
      "auxiliary_loss_mlp": 0.01051928,
      "balance_loss_clip": 1.07612252,
      "balance_loss_mlp": 1.03587031,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 1.906741283804164,
      "language_loss": 0.89395195,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.91709507,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.6889994144439697
    },
    {
      "auxiliary_loss_clip": 0.0125261,
      "auxiliary_loss_mlp": 0.01049811,
      "balance_loss_clip": 1.03467965,
      "balance_loss_mlp": 1.03468347,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.82472587480208,
      "language_loss": 0.78246236,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80548656,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.641174554824829
    },
    {
      "auxiliary_loss_clip": 0.01232452,
      "auxiliary_loss_mlp": 0.01051512,
      "balance_loss_clip": 0.98671293,
      "balance_loss_mlp": 1.03602684,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.740508514746703,
      "language_loss": 0.71725404,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.74009365,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.7622854709625244
    },
    {
      "auxiliary_loss_clip": 0.01239898,
      "auxiliary_loss_mlp": 0.01130518,
      "balance_loss_clip": 1.02908456,
      "balance_loss_mlp": 0.0,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 1.8678374994350269,
      "language_loss": 0.83726585,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.86097002,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.646418571472168
    },
    {
      "auxiliary_loss_clip": 0.01240666,
      "auxiliary_loss_mlp": 0.01039849,
      "balance_loss_clip": 0.91013682,
      "balance_loss_mlp": 1.02443576,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 2.9927270113359765,
      "language_loss": 0.74886096,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.77166605,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.8030455112457275
    },
    {
      "auxiliary_loss_clip": 0.01255534,
      "auxiliary_loss_mlp": 0.01045301,
      "balance_loss_clip": 1.03446829,
      "balance_loss_mlp": 1.02997112,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 2.5456778151423016,
      "language_loss": 0.81503236,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.83804077,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.689772605895996
    },
    {
      "auxiliary_loss_clip": 0.01255567,
      "auxiliary_loss_mlp": 0.01041229,
      "balance_loss_clip": 1.0723536,
      "balance_loss_mlp": 1.02642274,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 2.8300103198181814,
      "language_loss": 0.9066751,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.92964303,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.6343140602111816
    },
    {
      "auxiliary_loss_clip": 0.01251501,
      "auxiliary_loss_mlp": 0.01050207,
      "balance_loss_clip": 0.99467105,
      "balance_loss_mlp": 1.03511488,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 2.3749880169818036,
      "language_loss": 0.85370111,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.87671816,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.789907455444336
    },
    {
      "auxiliary_loss_clip": 0.01241055,
      "auxiliary_loss_mlp": 0.01044282,
      "balance_loss_clip": 0.91401339,
      "balance_loss_mlp": 1.0289402,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.329822135965108,
      "language_loss": 0.75783712,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.78069043,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.8316264152526855
    },
    {
      "auxiliary_loss_clip": 0.01243096,
      "auxiliary_loss_mlp": 0.01044933,
      "balance_loss_clip": 0.99144793,
      "balance_loss_mlp": 1.02924478,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 2.691394989173461,
      "language_loss": 0.94651353,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.96939379,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.6259140968322754
    },
    {
      "auxiliary_loss_clip": 0.01240613,
      "auxiliary_loss_mlp": 0.01047386,
      "balance_loss_clip": 0.91136652,
      "balance_loss_mlp": 1.03014779,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 2.045892110114207,
      "language_loss": 0.88600934,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.90888935,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.7525205612182617
    },
    {
      "auxiliary_loss_clip": 0.01239115,
      "auxiliary_loss_mlp": 0.01052194,
      "balance_loss_clip": 0.9125343,
      "balance_loss_mlp": 1.03623199,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 1.8966965843139063,
      "language_loss": 0.79866296,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.821576,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 2.7612383365631104
    },
    {
      "auxiliary_loss_clip": 0.01255298,
      "auxiliary_loss_mlp": 0.0105133,
      "balance_loss_clip": 1.03290212,
      "balance_loss_mlp": 1.0357132,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 1.957261481508997,
      "language_loss": 0.70697969,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.73004597,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.7628393173217773
    },
    {
      "auxiliary_loss_clip": 0.012474,
      "auxiliary_loss_mlp": 0.01046028,
      "balance_loss_clip": 0.95168865,
      "balance_loss_mlp": 1.03121078,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 1.9912865877617705,
      "language_loss": 0.8546716,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87760586,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 2.80507493019104
    },
    {
      "auxiliary_loss_clip": 0.01144495,
      "auxiliary_loss_mlp": 0.01009898,
      "balance_loss_clip": 0.92539823,
      "balance_loss_mlp": 0.99974138,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.9045075368720491,
      "language_loss": 0.61584312,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63738704,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 3.207118034362793
    },
    {
      "auxiliary_loss_clip": 0.01256812,
      "auxiliary_loss_mlp": 0.01051186,
      "balance_loss_clip": 0.99273092,
      "balance_loss_mlp": 1.03598654,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 2.0106448918804123,
      "language_loss": 0.92244738,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94552732,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 3.7556865215301514
    },
    {
      "auxiliary_loss_clip": 0.01255709,
      "auxiliary_loss_mlp": 0.0104785,
      "balance_loss_clip": 1.03414214,
      "balance_loss_mlp": 1.03243673,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 3.7583852898790377,
      "language_loss": 0.86345971,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88649529,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 2.649595022201538
    },
    {
      "auxiliary_loss_clip": 0.01237961,
      "auxiliary_loss_mlp": 0.01050651,
      "balance_loss_clip": 0.99161816,
      "balance_loss_mlp": 1.03446269,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 1.7217416596615387,
      "language_loss": 0.8051222,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.82800829,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 3.658647060394287
    },
    {
      "auxiliary_loss_clip": 0.01260421,
      "auxiliary_loss_mlp": 0.0104277,
      "balance_loss_clip": 1.03781819,
      "balance_loss_mlp": 1.02683151,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 3.011177624869204,
      "language_loss": 0.83836293,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86139482,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 2.68165922164917
    },
    {
      "auxiliary_loss_clip": 0.01257512,
      "auxiliary_loss_mlp": 0.01062426,
      "balance_loss_clip": 1.07405508,
      "balance_loss_mlp": 1.04715538,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 3.144600849464967,
      "language_loss": 0.83577573,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.85897505,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 3.5663681030273438
    },
    {
      "auxiliary_loss_clip": 0.01254782,
      "auxiliary_loss_mlp": 0.01046973,
      "balance_loss_clip": 1.03417552,
      "balance_loss_mlp": 1.03102231,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 2.034508138271821,
      "language_loss": 0.88272685,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.90574443,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 2.679504632949829
    },
    {
      "auxiliary_loss_clip": 0.01247955,
      "auxiliary_loss_mlp": 0.01049316,
      "balance_loss_clip": 0.91679609,
      "balance_loss_mlp": 1.03392577,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.1225522995214985,
      "language_loss": 0.91936934,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94234204,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.861133575439453
    },
    {
      "auxiliary_loss_clip": 0.01153248,
      "auxiliary_loss_mlp": 0.0101433,
      "balance_loss_clip": 1.04416871,
      "balance_loss_mlp": 1.00441194,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8519557461042494,
      "language_loss": 0.61241531,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63409114,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.354538917541504
    },
    {
      "auxiliary_loss_clip": 0.01245628,
      "auxiliary_loss_mlp": 0.01050476,
      "balance_loss_clip": 0.99231851,
      "balance_loss_mlp": 1.03453767,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 2.3161842312154683,
      "language_loss": 0.73592931,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.75889039,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.688232183456421
    },
    {
      "auxiliary_loss_clip": 0.01255835,
      "auxiliary_loss_mlp": 0.01060336,
      "balance_loss_clip": 0.99546152,
      "balance_loss_mlp": 1.04371798,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 3.6372491925157275,
      "language_loss": 0.82780719,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.85096896,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.6715714931488037
    },
    {
      "auxiliary_loss_clip": 0.01257711,
      "auxiliary_loss_mlp": 0.010542,
      "balance_loss_clip": 1.0752461,
      "balance_loss_mlp": 1.03901267,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 1.7238964610288765,
      "language_loss": 0.87852228,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90164137,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.6660726070404053
    },
    {
      "auxiliary_loss_clip": 0.01247986,
      "auxiliary_loss_mlp": 0.01047264,
      "balance_loss_clip": 0.98773199,
      "balance_loss_mlp": 1.03215981,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.7256852760841919,
      "language_loss": 0.84908098,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.87203348,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.7439754009246826
    },
    {
      "auxiliary_loss_clip": 0.01252211,
      "auxiliary_loss_mlp": 0.01050965,
      "balance_loss_clip": 1.03115582,
      "balance_loss_mlp": 1.03487182,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 2.6096668956968094,
      "language_loss": 0.74181527,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.76484698,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.7993547916412354
    },
    {
      "auxiliary_loss_clip": 0.01232824,
      "auxiliary_loss_mlp": 0.01052358,
      "balance_loss_clip": 0.91044074,
      "balance_loss_mlp": 1.03644395,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 2.1076136304895057,
      "language_loss": 0.73593485,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75878668,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.7895476818084717
    },
    {
      "auxiliary_loss_clip": 0.01146878,
      "auxiliary_loss_mlp": 0.01010303,
      "balance_loss_clip": 0.92641836,
      "balance_loss_mlp": 1.00043249,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 0.8435238857604066,
      "language_loss": 0.63683796,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65840983,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.373987913131714
    },
    {
      "auxiliary_loss_clip": 0.012591,
      "auxiliary_loss_mlp": 0.01056211,
      "balance_loss_clip": 0.99380374,
      "balance_loss_mlp": 1.03991544,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 1.856231341495713,
      "language_loss": 0.85209763,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.87525082,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.73726749420166
    },
    {
      "auxiliary_loss_clip": 0.01255009,
      "auxiliary_loss_mlp": 0.01047763,
      "balance_loss_clip": 1.03580475,
      "balance_loss_mlp": 1.03158665,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 3.273316395327301,
      "language_loss": 0.84815246,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87118018,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.6455228328704834
    },
    {
      "auxiliary_loss_clip": 0.01247732,
      "auxiliary_loss_mlp": 0.01055754,
      "balance_loss_clip": 0.99355543,
      "balance_loss_mlp": 1.04034042,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 2.321131923394081,
      "language_loss": 0.76644957,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.78948444,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.7093756198883057
    },
    {
      "auxiliary_loss_clip": 0.01253165,
      "auxiliary_loss_mlp": 0.0104397,
      "balance_loss_clip": 1.06942701,
      "balance_loss_mlp": 1.02976036,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 1.584824046273688,
      "language_loss": 0.78517783,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.80814916,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.6518614292144775
    },
    {
      "auxiliary_loss_clip": 0.01253322,
      "auxiliary_loss_mlp": 0.01042385,
      "balance_loss_clip": 0.98947412,
      "balance_loss_mlp": 1.02681661,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.4100125770900553,
      "language_loss": 0.79622412,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.8191812,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.7635838985443115
    },
    {
      "auxiliary_loss_clip": 0.01255048,
      "auxiliary_loss_mlp": 0.01053106,
      "balance_loss_clip": 0.99701595,
      "balance_loss_mlp": 1.03808618,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 1.9010819279425812,
      "language_loss": 0.84478509,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86786664,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.7813637256622314
    },
    {
      "auxiliary_loss_clip": 0.01250199,
      "auxiliary_loss_mlp": 0.01130503,
      "balance_loss_clip": 0.99309343,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 2.609196622658837,
      "language_loss": 0.68372083,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.70752788,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.66489315032959
    },
    {
      "auxiliary_loss_clip": 0.01240267,
      "auxiliary_loss_mlp": 0.01130598,
      "balance_loss_clip": 0.84435332,
      "balance_loss_mlp": 0.0,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 1.670310442572063,
      "language_loss": 0.83749413,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.86120272,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 3.002549886703491
    },
    {
      "auxiliary_loss_clip": 0.01248243,
      "auxiliary_loss_mlp": 0.01049551,
      "balance_loss_clip": 1.03154588,
      "balance_loss_mlp": 1.0347209,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 2.1653838550323137,
      "language_loss": 0.83664227,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.85962021,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 2.931641101837158
    },
    {
      "auxiliary_loss_clip": 0.012422,
      "auxiliary_loss_mlp": 0.01053836,
      "balance_loss_clip": 0.94908774,
      "balance_loss_mlp": 1.0388757,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 1.7954678338581713,
      "language_loss": 0.88325846,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.90621877,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 2.856818437576294
    },
    {
      "auxiliary_loss_clip": 0.0124835,
      "auxiliary_loss_mlp": 0.01052618,
      "balance_loss_clip": 0.99395835,
      "balance_loss_mlp": 1.03821731,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 2.173499740348231,
      "language_loss": 0.81018376,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83319342,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 2.7046563625335693
    },
    {
      "auxiliary_loss_clip": 0.01249066,
      "auxiliary_loss_mlp": 0.01052101,
      "balance_loss_clip": 1.02857304,
      "balance_loss_mlp": 1.03720009,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.8263374415742801,
      "language_loss": 0.74060404,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76361573,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 3.7567479610443115
    },
    {
      "auxiliary_loss_clip": 0.01242926,
      "auxiliary_loss_mlp": 0.01046971,
      "balance_loss_clip": 0.99063075,
      "balance_loss_mlp": 1.03156924,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 2.013603858790762,
      "language_loss": 0.78056836,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80346727,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 3.609637498855591
    },
    {
      "auxiliary_loss_clip": 0.01248395,
      "auxiliary_loss_mlp": 0.01050102,
      "balance_loss_clip": 1.02669501,
      "balance_loss_mlp": 1.03456903,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 2.2458340844521087,
      "language_loss": 0.831779,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.85476398,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 3.6277496814727783
    },
    {
      "auxiliary_loss_clip": 0.01246194,
      "auxiliary_loss_mlp": 0.01046353,
      "balance_loss_clip": 0.99152493,
      "balance_loss_mlp": 1.03180957,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 1.7613133290650316,
      "language_loss": 0.81293166,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83585715,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 2.6526918411254883
    },
    {
      "auxiliary_loss_clip": 0.01254672,
      "auxiliary_loss_mlp": 0.01049516,
      "balance_loss_clip": 1.03206825,
      "balance_loss_mlp": 1.03437662,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 3.6024275758816935,
      "language_loss": 0.83441699,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85745883,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 3.543985366821289
    },
    {
      "auxiliary_loss_clip": 0.01252987,
      "auxiliary_loss_mlp": 0.01130246,
      "balance_loss_clip": 1.03361809,
      "balance_loss_mlp": 0.0,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 1.9959035536057423,
      "language_loss": 0.90832114,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.93215346,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 2.644397735595703
    },
    {
      "auxiliary_loss_clip": 0.01254533,
      "auxiliary_loss_mlp": 0.01050541,
      "balance_loss_clip": 1.03184271,
      "balance_loss_mlp": 1.03523493,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 2.117126362659637,
      "language_loss": 0.75473833,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.77778906,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.6573104858398438
    },
    {
      "auxiliary_loss_clip": 0.01252735,
      "auxiliary_loss_mlp": 0.0104709,
      "balance_loss_clip": 1.07065976,
      "balance_loss_mlp": 1.03297555,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.7813715852048297,
      "language_loss": 0.78727925,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.81027746,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.7306854724884033
    },
    {
      "auxiliary_loss_clip": 0.01233891,
      "auxiliary_loss_mlp": 0.0113034,
      "balance_loss_clip": 0.98819697,
      "balance_loss_mlp": 0.0,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.734287525277213,
      "language_loss": 0.86471748,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.88835979,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.744828701019287
    },
    {
      "auxiliary_loss_clip": 0.01250994,
      "auxiliary_loss_mlp": 0.01045942,
      "balance_loss_clip": 1.02849078,
      "balance_loss_mlp": 1.03088617,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 2.3309961628826574,
      "language_loss": 0.9040345,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.92700386,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.6303975582122803
    },
    {
      "auxiliary_loss_clip": 0.01246818,
      "auxiliary_loss_mlp": 0.01045359,
      "balance_loss_clip": 0.99304372,
      "balance_loss_mlp": 1.03073156,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.180734393546764,
      "language_loss": 0.83980823,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86272997,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.7691404819488525
    },
    {
      "auxiliary_loss_clip": 0.01235111,
      "auxiliary_loss_mlp": 0.01045698,
      "balance_loss_clip": 0.94875741,
      "balance_loss_mlp": 1.03065348,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 21.001508234440724,
      "language_loss": 0.83298492,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.855793,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.7930033206939697
    },
    {
      "auxiliary_loss_clip": 0.01241074,
      "auxiliary_loss_mlp": 0.01052739,
      "balance_loss_clip": 0.95006895,
      "balance_loss_mlp": 1.03774214,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 2.9453008077780023,
      "language_loss": 0.75736082,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.78029895,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.6678123474121094
    },
    {
      "auxiliary_loss_clip": 0.01233082,
      "auxiliary_loss_mlp": 0.01041155,
      "balance_loss_clip": 0.9864102,
      "balance_loss_mlp": 1.02646828,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.08024967472353,
      "language_loss": 0.8957473,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.91848969,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.673401355743408
    },
    {
      "auxiliary_loss_clip": 0.01254186,
      "auxiliary_loss_mlp": 0.01044668,
      "balance_loss_clip": 1.06909072,
      "balance_loss_mlp": 1.02926588,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 2.4344127336738213,
      "language_loss": 0.76393461,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.78692317,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.621530771255493
    },
    {
      "auxiliary_loss_clip": 0.01255404,
      "auxiliary_loss_mlp": 0.01130318,
      "balance_loss_clip": 1.03612804,
      "balance_loss_mlp": 0.0,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 1.6916319850765733,
      "language_loss": 0.75828207,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.78213924,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.7008626461029053
    },
    {
      "auxiliary_loss_clip": 0.01252819,
      "auxiliary_loss_mlp": 0.01044596,
      "balance_loss_clip": 1.0325675,
      "balance_loss_mlp": 1.02995718,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 1.8883916128476534,
      "language_loss": 0.71590757,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.73888171,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.784925937652588
    },
    {
      "auxiliary_loss_clip": 0.01243153,
      "auxiliary_loss_mlp": 0.01043611,
      "balance_loss_clip": 0.9886601,
      "balance_loss_mlp": 1.02882886,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 1.9030119181031004,
      "language_loss": 0.92401731,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.94688493,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.704291582107544
    },
    {
      "auxiliary_loss_clip": 0.01240178,
      "auxiliary_loss_mlp": 0.01052403,
      "balance_loss_clip": 1.02693129,
      "balance_loss_mlp": 1.03683472,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 4.0559671133046695,
      "language_loss": 0.69303733,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.71596313,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.662444591522217
    },
    {
      "auxiliary_loss_clip": 0.01257523,
      "auxiliary_loss_mlp": 0.01051864,
      "balance_loss_clip": 0.99325514,
      "balance_loss_mlp": 1.0364027,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.4358204833838917,
      "language_loss": 0.74860466,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77169847,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.700622797012329
    },
    {
      "auxiliary_loss_clip": 0.01252635,
      "auxiliary_loss_mlp": 0.01046123,
      "balance_loss_clip": 1.03237474,
      "balance_loss_mlp": 1.03032792,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 2.2356279364518388,
      "language_loss": 0.74356705,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76655465,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.633592367172241
    },
    {
      "auxiliary_loss_clip": 0.01238387,
      "auxiliary_loss_mlp": 0.01038993,
      "balance_loss_clip": 1.02963376,
      "balance_loss_mlp": 1.0241518,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 2.412995429975761,
      "language_loss": 0.86209536,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88486916,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.751154661178589
    },
    {
      "auxiliary_loss_clip": 0.0124624,
      "auxiliary_loss_mlp": 0.01043824,
      "balance_loss_clip": 0.95446193,
      "balance_loss_mlp": 1.02899468,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 2.537680092678929,
      "language_loss": 0.80142975,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.82433039,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.7100913524627686
    },
    {
      "auxiliary_loss_clip": 0.01236532,
      "auxiliary_loss_mlp": 0.01045021,
      "balance_loss_clip": 1.02631402,
      "balance_loss_mlp": 1.02972674,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 3.5916605675453113,
      "language_loss": 0.85755789,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88037336,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 2.6777496337890625
    },
    {
      "auxiliary_loss_clip": 0.01246193,
      "auxiliary_loss_mlp": 0.01049918,
      "balance_loss_clip": 0.99264359,
      "balance_loss_mlp": 1.0342772,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 2.087863814237613,
      "language_loss": 0.8353864,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85834742,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 2.8323965072631836
    },
    {
      "auxiliary_loss_clip": 0.01245519,
      "auxiliary_loss_mlp": 0.01130734,
      "balance_loss_clip": 0.99178088,
      "balance_loss_mlp": 0.0,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.2695144847182744,
      "language_loss": 0.85732293,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.8810854,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 2.7144289016723633
    },
    {
      "auxiliary_loss_clip": 0.01236074,
      "auxiliary_loss_mlp": 0.01057176,
      "balance_loss_clip": 0.9145624,
      "balance_loss_mlp": 1.04167879,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 4.068568147357138,
      "language_loss": 0.85351712,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.87644964,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 2.8065648078918457
    },
    {
      "auxiliary_loss_clip": 0.01248846,
      "auxiliary_loss_mlp": 0.01045188,
      "balance_loss_clip": 1.03237784,
      "balance_loss_mlp": 1.03087139,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 1.7867626302017703,
      "language_loss": 0.79042244,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81336284,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 4.587406158447266
    },
    {
      "auxiliary_loss_clip": 0.01248033,
      "auxiliary_loss_mlp": 0.01048086,
      "balance_loss_clip": 0.99184728,
      "balance_loss_mlp": 1.03298259,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 2.3161337914848015,
      "language_loss": 0.75297612,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.77593732,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 3.633558511734009
    },
    {
      "auxiliary_loss_clip": 0.0123935,
      "auxiliary_loss_mlp": 0.01041437,
      "balance_loss_clip": 0.91454589,
      "balance_loss_mlp": 1.02731109,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 2.3998670557862365,
      "language_loss": 0.81047142,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83327937,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 2.8365273475646973
    },
    {
      "auxiliary_loss_clip": 0.01251197,
      "auxiliary_loss_mlp": 0.01129912,
      "balance_loss_clip": 0.99578011,
      "balance_loss_mlp": 0.0,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 2.060213636269515,
      "language_loss": 0.84864712,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.87245822,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 3.726989507675171
    },
    {
      "auxiliary_loss_clip": 0.01247891,
      "auxiliary_loss_mlp": 0.01047522,
      "balance_loss_clip": 0.9902187,
      "balance_loss_mlp": 1.03196478,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 2.049842151656829,
      "language_loss": 0.80917114,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.83212525,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 2.679940700531006
    },
    {
      "auxiliary_loss_clip": 0.01237388,
      "auxiliary_loss_mlp": 0.01040936,
      "balance_loss_clip": 0.99124902,
      "balance_loss_mlp": 1.02627325,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 2.2371465133863153,
      "language_loss": 0.9023838,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92516708,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.6802895069122314
    },
    {
      "auxiliary_loss_clip": 0.01257881,
      "auxiliary_loss_mlp": 0.01052777,
      "balance_loss_clip": 1.07269835,
      "balance_loss_mlp": 1.03816211,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 1.984426427307951,
      "language_loss": 0.91019911,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.93330562,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.7023134231567383
    },
    {
      "auxiliary_loss_clip": 0.01238588,
      "auxiliary_loss_mlp": 0.01046265,
      "balance_loss_clip": 0.98821729,
      "balance_loss_mlp": 1.03230619,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 1.6452481127848406,
      "language_loss": 0.83564621,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85849476,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.7573189735412598
    },
    {
      "auxiliary_loss_clip": 0.01253867,
      "auxiliary_loss_mlp": 0.01059521,
      "balance_loss_clip": 0.99529308,
      "balance_loss_mlp": 1.04469121,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 2.8146380477697317,
      "language_loss": 0.66956866,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.69270253,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.729935646057129
    },
    {
      "auxiliary_loss_clip": 0.01251095,
      "auxiliary_loss_mlp": 0.01042877,
      "balance_loss_clip": 1.03095424,
      "balance_loss_mlp": 1.02881002,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 1.541101962487206,
      "language_loss": 0.79845917,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82139885,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.7694082260131836
    },
    {
      "auxiliary_loss_clip": 0.01251744,
      "auxiliary_loss_mlp": 0.01039332,
      "balance_loss_clip": 1.03310251,
      "balance_loss_mlp": 1.02531302,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.515220709318716,
      "language_loss": 0.81818485,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.84109557,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.661341428756714
    },
    {
      "auxiliary_loss_clip": 0.01252698,
      "auxiliary_loss_mlp": 0.01048114,
      "balance_loss_clip": 1.03028226,
      "balance_loss_mlp": 1.03316474,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 1.8659039298054234,
      "language_loss": 0.84923643,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87224454,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.743417978286743
    },
    {
      "auxiliary_loss_clip": 0.01245287,
      "auxiliary_loss_mlp": 0.010409,
      "balance_loss_clip": 1.02984643,
      "balance_loss_mlp": 1.02642846,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 1.8523473113564572,
      "language_loss": 0.85359859,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87646043,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.7132906913757324
    },
    {
      "auxiliary_loss_clip": 0.01253651,
      "auxiliary_loss_mlp": 0.0105475,
      "balance_loss_clip": 1.03243613,
      "balance_loss_mlp": 1.0410533,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 1.9654895863581325,
      "language_loss": 0.84961027,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.87269431,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.7591214179992676
    },
    {
      "auxiliary_loss_clip": 0.01239973,
      "auxiliary_loss_mlp": 0.01048111,
      "balance_loss_clip": 0.95026577,
      "balance_loss_mlp": 1.03440166,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 2.718417706687038,
      "language_loss": 0.74556088,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.76844168,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.8216373920440674
    },
    {
      "auxiliary_loss_clip": 0.01251945,
      "auxiliary_loss_mlp": 0.01041681,
      "balance_loss_clip": 1.03123379,
      "balance_loss_mlp": 1.02683914,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 2.751308918354223,
      "language_loss": 0.66412938,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68706566,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.7109463214874268
    },
    {
      "auxiliary_loss_clip": 0.01224213,
      "auxiliary_loss_mlp": 0.01042771,
      "balance_loss_clip": 0.94329393,
      "balance_loss_mlp": 1.02882314,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.4497064364982,
      "language_loss": 0.89800489,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92067468,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.800931692123413
    },
    {
      "auxiliary_loss_clip": 0.0125357,
      "auxiliary_loss_mlp": 0.01047321,
      "balance_loss_clip": 1.07077384,
      "balance_loss_mlp": 1.03323686,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.239894129787103,
      "language_loss": 0.7734617,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.79647064,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.6357569694519043
    },
    {
      "auxiliary_loss_clip": 0.01245071,
      "auxiliary_loss_mlp": 0.01046589,
      "balance_loss_clip": 0.99397135,
      "balance_loss_mlp": 1.03242695,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 2.4756924013568495,
      "language_loss": 0.77848256,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80139923,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.824266195297241
    },
    {
      "auxiliary_loss_clip": 0.01246348,
      "auxiliary_loss_mlp": 0.01041699,
      "balance_loss_clip": 0.98950446,
      "balance_loss_mlp": 1.02732253,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 1.863625092418395,
      "language_loss": 0.75953114,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78241163,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.7658772468566895
    },
    {
      "auxiliary_loss_clip": 0.01241403,
      "auxiliary_loss_mlp": 0.01129479,
      "balance_loss_clip": 0.98687899,
      "balance_loss_mlp": 0.0,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 1.8115832499577809,
      "language_loss": 0.80954152,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.8332504,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.7647695541381836
    },
    {
      "auxiliary_loss_clip": 0.01231196,
      "auxiliary_loss_mlp": 0.01048989,
      "balance_loss_clip": 0.99125206,
      "balance_loss_mlp": 1.03443336,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 3.4975047562422548,
      "language_loss": 0.78755462,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.81035638,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.708238124847412
    },
    {
      "auxiliary_loss_clip": 0.01237741,
      "auxiliary_loss_mlp": 0.01044368,
      "balance_loss_clip": 0.91518092,
      "balance_loss_mlp": 1.03028965,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 2.183684652389814,
      "language_loss": 0.58399874,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.60681981,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 2.7953338623046875
    },
    {
      "auxiliary_loss_clip": 0.01235958,
      "auxiliary_loss_mlp": 0.01044667,
      "balance_loss_clip": 1.02611113,
      "balance_loss_mlp": 1.02965856,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.710601677682532,
      "language_loss": 0.85167271,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87447894,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 2.8089675903320312
    },
    {
      "auxiliary_loss_clip": 0.01248527,
      "auxiliary_loss_mlp": 0.0103789,
      "balance_loss_clip": 1.02804017,
      "balance_loss_mlp": 1.02401948,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 2.499571492883242,
      "language_loss": 0.71357405,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.73643827,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 2.6019771099090576
    },
    {
      "auxiliary_loss_clip": 0.01249576,
      "auxiliary_loss_mlp": 0.01130047,
      "balance_loss_clip": 1.03075862,
      "balance_loss_mlp": 0.0,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 1.5833303335005169,
      "language_loss": 0.82512015,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.84891641,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 2.7547402381896973
    },
    {
      "auxiliary_loss_clip": 0.01244354,
      "auxiliary_loss_mlp": 0.01045025,
      "balance_loss_clip": 0.99603009,
      "balance_loss_mlp": 1.03179884,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.0453250324250627,
      "language_loss": 0.86590427,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88879812,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 3.6292178630828857
    },
    {
      "auxiliary_loss_clip": 0.01250252,
      "auxiliary_loss_mlp": 0.01044743,
      "balance_loss_clip": 1.06624484,
      "balance_loss_mlp": 1.03039062,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.3532175220348233,
      "language_loss": 0.75960958,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.78255951,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 3.704881429672241
    },
    {
      "auxiliary_loss_clip": 0.01214699,
      "auxiliary_loss_mlp": 0.01038991,
      "balance_loss_clip": 0.90744913,
      "balance_loss_mlp": 1.02521014,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 1.97037677740796,
      "language_loss": 0.91455787,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93709481,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 3.6932387351989746
    },
    {
      "auxiliary_loss_clip": 0.01228522,
      "auxiliary_loss_mlp": 0.01041902,
      "balance_loss_clip": 0.86690605,
      "balance_loss_mlp": 1.02707839,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 1.7709309045179566,
      "language_loss": 0.82928073,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.85198498,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 3.8824379444122314
    },
    {
      "auxiliary_loss_clip": 0.0124058,
      "auxiliary_loss_mlp": 0.01044003,
      "balance_loss_clip": 0.98758632,
      "balance_loss_mlp": 1.02946508,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.2165373762791347,
      "language_loss": 0.7861312,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.80897707,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 3.0365660190582275
    },
    {
      "auxiliary_loss_clip": 0.01245575,
      "auxiliary_loss_mlp": 0.01044906,
      "balance_loss_clip": 0.99256122,
      "balance_loss_mlp": 1.03014755,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.0338482780269307,
      "language_loss": 0.77025455,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.79315937,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.7360289096832275
    },
    {
      "auxiliary_loss_clip": 0.01239632,
      "auxiliary_loss_mlp": 0.01042915,
      "balance_loss_clip": 0.95197916,
      "balance_loss_mlp": 1.02940273,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 2.7061830362800667,
      "language_loss": 0.77484298,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.7976684,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.6785061359405518
    },
    {
      "auxiliary_loss_clip": 0.0124732,
      "auxiliary_loss_mlp": 0.01052482,
      "balance_loss_clip": 1.02875209,
      "balance_loss_mlp": 1.03828454,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 1.7730560531804918,
      "language_loss": 0.81175625,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83475423,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.6967084407806396
    },
    {
      "auxiliary_loss_clip": 0.01243888,
      "auxiliary_loss_mlp": 0.01042549,
      "balance_loss_clip": 0.95259035,
      "balance_loss_mlp": 1.0282979,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 1.6928442202763552,
      "language_loss": 0.84572732,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.86859167,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.713872194290161
    },
    {
      "auxiliary_loss_clip": 0.01242613,
      "auxiliary_loss_mlp": 0.01046705,
      "balance_loss_clip": 1.02792895,
      "balance_loss_mlp": 1.03217888,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 1.824754562763651,
      "language_loss": 0.86743021,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.8903234,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.672173261642456
    },
    {
      "auxiliary_loss_clip": 0.01145391,
      "auxiliary_loss_mlp": 0.01011729,
      "balance_loss_clip": 0.89462614,
      "balance_loss_mlp": 1.00295568,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.7958578808109096,
      "language_loss": 0.54436463,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.56593585,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.271183967590332
    },
    {
      "auxiliary_loss_clip": 0.01246068,
      "auxiliary_loss_mlp": 0.01045978,
      "balance_loss_clip": 1.06653666,
      "balance_loss_mlp": 1.03200638,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 2.1076672699133505,
      "language_loss": 0.79001701,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.8129375,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.641880512237549
    },
    {
      "auxiliary_loss_clip": 0.01226641,
      "auxiliary_loss_mlp": 0.01036885,
      "balance_loss_clip": 0.98499364,
      "balance_loss_mlp": 1.02191246,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 2.0440893650682943,
      "language_loss": 0.75504887,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77768415,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.708059549331665
    },
    {
      "auxiliary_loss_clip": 0.01219869,
      "auxiliary_loss_mlp": 0.01047248,
      "balance_loss_clip": 0.90839827,
      "balance_loss_mlp": 1.03272796,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 2.064570531699201,
      "language_loss": 0.83378637,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85645753,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.750368118286133
    },
    {
      "auxiliary_loss_clip": 0.01247961,
      "auxiliary_loss_mlp": 0.01049307,
      "balance_loss_clip": 1.06767559,
      "balance_loss_mlp": 1.03494823,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 3.5021988314511585,
      "language_loss": 1.05998826,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.08296084,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.618579149246216
    },
    {
      "auxiliary_loss_clip": 0.01243522,
      "auxiliary_loss_mlp": 0.01042314,
      "balance_loss_clip": 1.03094172,
      "balance_loss_mlp": 1.02752566,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 1.8070869462722337,
      "language_loss": 0.80440164,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.82725996,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.704911708831787
    },
    {
      "auxiliary_loss_clip": 0.01239786,
      "auxiliary_loss_mlp": 0.01050135,
      "balance_loss_clip": 0.91370583,
      "balance_loss_mlp": 1.03548455,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.163744326538849,
      "language_loss": 0.75973421,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78263336,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.7964820861816406
    },
    {
      "auxiliary_loss_clip": 0.0125015,
      "auxiliary_loss_mlp": 0.0105093,
      "balance_loss_clip": 0.98913932,
      "balance_loss_mlp": 1.03775132,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 2.302871240270919,
      "language_loss": 0.72189856,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74490935,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.683856964111328
    },
    {
      "auxiliary_loss_clip": 0.01248367,
      "auxiliary_loss_mlp": 0.01045445,
      "balance_loss_clip": 1.0696882,
      "balance_loss_mlp": 1.03114557,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 2.074475007140264,
      "language_loss": 0.75528991,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.77822804,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.684900999069214
    },
    {
      "auxiliary_loss_clip": 0.01247351,
      "auxiliary_loss_mlp": 0.01046976,
      "balance_loss_clip": 1.06509805,
      "balance_loss_mlp": 1.03346944,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 2.252076363588934,
      "language_loss": 0.71640122,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73934448,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.6805202960968018
    },
    {
      "auxiliary_loss_clip": 0.01244501,
      "auxiliary_loss_mlp": 0.01056659,
      "balance_loss_clip": 0.95565182,
      "balance_loss_mlp": 1.04228199,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 1.767337246012119,
      "language_loss": 0.79468977,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81770134,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.706228494644165
    },
    {
      "auxiliary_loss_clip": 0.01245654,
      "auxiliary_loss_mlp": 0.01041548,
      "balance_loss_clip": 1.06749821,
      "balance_loss_mlp": 1.02650356,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 2.4058051980413686,
      "language_loss": 0.90425599,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92712802,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.736837863922119
    },
    {
      "auxiliary_loss_clip": 0.01252411,
      "auxiliary_loss_mlp": 0.01050873,
      "balance_loss_clip": 1.03063619,
      "balance_loss_mlp": 1.03672314,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 2.227565454039306,
      "language_loss": 0.81531298,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83834589,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 2.6587626934051514
    },
    {
      "auxiliary_loss_clip": 0.0124824,
      "auxiliary_loss_mlp": 0.01051183,
      "balance_loss_clip": 0.99074817,
      "balance_loss_mlp": 1.03711605,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 2.1108848749879137,
      "language_loss": 0.82833123,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.85132545,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 2.69197416305542
    },
    {
      "auxiliary_loss_clip": 0.01146966,
      "auxiliary_loss_mlp": 0.01027563,
      "balance_loss_clip": 0.92807651,
      "balance_loss_mlp": 1.01855087,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.05704244510929,
      "language_loss": 0.70184135,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.72358668,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 3.0979115962982178
    },
    {
      "auxiliary_loss_clip": 0.01248623,
      "auxiliary_loss_mlp": 0.01049343,
      "balance_loss_clip": 1.06797862,
      "balance_loss_mlp": 1.03567028,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 1.7945043937549885,
      "language_loss": 0.77109873,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.79407841,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 2.704047441482544
    },
    {
      "auxiliary_loss_clip": 0.01240585,
      "auxiliary_loss_mlp": 0.01047815,
      "balance_loss_clip": 0.99142361,
      "balance_loss_mlp": 1.03330684,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.0760090243398577,
      "language_loss": 0.83619976,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.85908377,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 3.8053674697875977
    },
    {
      "auxiliary_loss_clip": 0.01248124,
      "auxiliary_loss_mlp": 0.01040829,
      "balance_loss_clip": 1.02810907,
      "balance_loss_mlp": 1.02605939,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 2.6699388176487027,
      "language_loss": 0.72378445,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.746674,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 3.7215278148651123
    },
    {
      "auxiliary_loss_clip": 0.01242135,
      "auxiliary_loss_mlp": 0.01040388,
      "balance_loss_clip": 0.9908917,
      "balance_loss_mlp": 1.02627909,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 1.8007452388397582,
      "language_loss": 0.80109119,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82391649,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 3.754164695739746
    },
    {
      "auxiliary_loss_clip": 0.01253245,
      "auxiliary_loss_mlp": 0.01048694,
      "balance_loss_clip": 1.03085935,
      "balance_loss_mlp": 1.03409052,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 1.7627481356336043,
      "language_loss": 0.76709759,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.79011697,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 3.7267701625823975
    },
    {
      "auxiliary_loss_clip": 0.01245361,
      "auxiliary_loss_mlp": 0.01048793,
      "balance_loss_clip": 0.99158794,
      "balance_loss_mlp": 1.03391528,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 1.577981168155191,
      "language_loss": 0.79899681,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.82193834,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 2.757697343826294
    },
    {
      "auxiliary_loss_clip": 0.0124426,
      "auxiliary_loss_mlp": 0.01042474,
      "balance_loss_clip": 0.98906207,
      "balance_loss_mlp": 1.02753663,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 3.2035282256218163,
      "language_loss": 0.71758127,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.74044859,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.8257834911346436
    },
    {
      "auxiliary_loss_clip": 0.01246484,
      "auxiliary_loss_mlp": 0.01047398,
      "balance_loss_clip": 1.06678128,
      "balance_loss_mlp": 1.03318787,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 1.8962285418291118,
      "language_loss": 0.77944732,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80238616,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.7548158168792725
    },
    {
      "auxiliary_loss_clip": 0.01247784,
      "auxiliary_loss_mlp": 0.0113003,
      "balance_loss_clip": 1.06844544,
      "balance_loss_mlp": 0.0,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.2380353491370943,
      "language_loss": 0.88073039,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.90450847,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.664882183074951
    },
    {
      "auxiliary_loss_clip": 0.01245976,
      "auxiliary_loss_mlp": 0.0104575,
      "balance_loss_clip": 0.99067402,
      "balance_loss_mlp": 1.03210068,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 2.4192516456156734,
      "language_loss": 0.80025995,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.82317722,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.6762990951538086
    },
    {
      "auxiliary_loss_clip": 0.01253427,
      "auxiliary_loss_mlp": 0.01041839,
      "balance_loss_clip": 1.03075361,
      "balance_loss_mlp": 1.0276413,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 2.3183304877775766,
      "language_loss": 0.75335842,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77631116,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.675182819366455
    },
    {
      "auxiliary_loss_clip": 0.01250247,
      "auxiliary_loss_mlp": 0.01039529,
      "balance_loss_clip": 1.07132149,
      "balance_loss_mlp": 1.02519965,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.60373254058056,
      "language_loss": 0.84564114,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.86853886,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.65260910987854
    },
    {
      "auxiliary_loss_clip": 0.01242162,
      "auxiliary_loss_mlp": 0.01044228,
      "balance_loss_clip": 0.94820046,
      "balance_loss_mlp": 1.02862394,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 2.329103420404068,
      "language_loss": 0.86265731,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88552123,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.7189724445343018
    },
    {
      "auxiliary_loss_clip": 0.01250594,
      "auxiliary_loss_mlp": 0.01130212,
      "balance_loss_clip": 1.06900382,
      "balance_loss_mlp": 0.0,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 2.3493010361189133,
      "language_loss": 0.87316108,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.8969692,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.66332745552063
    },
    {
      "auxiliary_loss_clip": 0.01251172,
      "auxiliary_loss_mlp": 0.01049691,
      "balance_loss_clip": 0.9936353,
      "balance_loss_mlp": 1.03585112,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 1.8391963753352272,
      "language_loss": 0.76435769,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.78736627,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.7626030445098877
    },
    {
      "auxiliary_loss_clip": 0.01253967,
      "auxiliary_loss_mlp": 0.01057633,
      "balance_loss_clip": 1.07292569,
      "balance_loss_mlp": 1.04380488,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 1.9204877627916774,
      "language_loss": 0.85791409,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88103008,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.6892714500427246
    },
    {
      "auxiliary_loss_clip": 0.01248754,
      "auxiliary_loss_mlp": 0.01045666,
      "balance_loss_clip": 0.99498451,
      "balance_loss_mlp": 1.03146863,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.288443770192401,
      "language_loss": 0.79878461,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82172883,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.7007129192352295
    },
    {
      "auxiliary_loss_clip": 0.01244205,
      "auxiliary_loss_mlp": 0.01046507,
      "balance_loss_clip": 1.0285573,
      "balance_loss_mlp": 1.03117657,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.6637636564174243,
      "language_loss": 0.87865698,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90156412,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.7927794456481934
    },
    {
      "auxiliary_loss_clip": 0.011546,
      "auxiliary_loss_mlp": 0.01008784,
      "balance_loss_clip": 1.00657403,
      "balance_loss_mlp": 0.99958116,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9575458666651306,
      "language_loss": 0.64492768,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66656154,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.224428653717041
    },
    {
      "auxiliary_loss_clip": 0.01248345,
      "auxiliary_loss_mlp": 0.01041919,
      "balance_loss_clip": 1.02728796,
      "balance_loss_mlp": 1.02737498,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 2.766035128272467,
      "language_loss": 0.80528909,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.82819176,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.645012378692627
    },
    {
      "auxiliary_loss_clip": 0.0124668,
      "auxiliary_loss_mlp": 0.01129725,
      "balance_loss_clip": 1.03222227,
      "balance_loss_mlp": 0.0,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 4.357399354975338,
      "language_loss": 0.81718123,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.84094536,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.739753007888794
    },
    {
      "auxiliary_loss_clip": 0.01248194,
      "auxiliary_loss_mlp": 0.01038326,
      "balance_loss_clip": 1.03144598,
      "balance_loss_mlp": 1.02385378,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 2.639026384317663,
      "language_loss": 0.86152947,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.88439465,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.7865912914276123
    },
    {
      "auxiliary_loss_clip": 0.01240358,
      "auxiliary_loss_mlp": 0.01047031,
      "balance_loss_clip": 1.02735758,
      "balance_loss_mlp": 1.03265405,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.5036435572136866,
      "language_loss": 0.90996075,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93283463,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.7202258110046387
    },
    {
      "auxiliary_loss_clip": 0.01257217,
      "auxiliary_loss_mlp": 0.01043984,
      "balance_loss_clip": 1.03765464,
      "balance_loss_mlp": 1.02977419,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 2.86522439429699,
      "language_loss": 0.7889694,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81198138,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.632133722305298
    },
    {
      "auxiliary_loss_clip": 0.01248864,
      "auxiliary_loss_mlp": 0.01041414,
      "balance_loss_clip": 1.07116389,
      "balance_loss_mlp": 1.02771688,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 1.9378852865396665,
      "language_loss": 0.77259821,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79550099,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 2.683227300643921
    },
    {
      "auxiliary_loss_clip": 0.01247164,
      "auxiliary_loss_mlp": 0.01052133,
      "balance_loss_clip": 0.95212662,
      "balance_loss_mlp": 1.03730321,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 1.8335074037753851,
      "language_loss": 0.92157614,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.94456911,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 2.738724946975708
    },
    {
      "auxiliary_loss_clip": 0.01250753,
      "auxiliary_loss_mlp": 0.01045961,
      "balance_loss_clip": 1.06998551,
      "balance_loss_mlp": 1.03163195,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.238294958058628,
      "language_loss": 0.827075,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.85004216,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 2.6504878997802734
    },
    {
      "auxiliary_loss_clip": 0.01225771,
      "auxiliary_loss_mlp": 0.01043126,
      "balance_loss_clip": 0.98731256,
      "balance_loss_mlp": 1.02870131,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 1.6986875426986112,
      "language_loss": 0.88102007,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90370905,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 2.7229909896850586
    },
    {
      "auxiliary_loss_clip": 0.0114079,
      "auxiliary_loss_mlp": 0.01015473,
      "balance_loss_clip": 0.84814823,
      "balance_loss_mlp": 1.00660419,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.9291605996327463,
      "language_loss": 0.60503703,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62659967,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 5.120313405990601
    },
    {
      "auxiliary_loss_clip": 0.01237677,
      "auxiliary_loss_mlp": 0.01041736,
      "balance_loss_clip": 0.94968426,
      "balance_loss_mlp": 1.02767491,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 1.9539797640185805,
      "language_loss": 0.82879347,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85158759,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 3.0023744106292725
    },
    {
      "auxiliary_loss_clip": 0.01225066,
      "auxiliary_loss_mlp": 0.01129922,
      "balance_loss_clip": 0.94663274,
      "balance_loss_mlp": 0.0,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.9608649697238376,
      "language_loss": 0.84028757,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.86383748,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 3.7418384552001953
    },
    {
      "auxiliary_loss_clip": 0.01237672,
      "auxiliary_loss_mlp": 0.01043831,
      "balance_loss_clip": 0.94630694,
      "balance_loss_mlp": 1.03020525,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 1.8756015515114843,
      "language_loss": 0.77506286,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79787791,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.70637583732605
    },
    {
      "auxiliary_loss_clip": 0.0125007,
      "auxiliary_loss_mlp": 0.01043836,
      "balance_loss_clip": 0.99365222,
      "balance_loss_mlp": 1.02882802,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 3.5440928732593147,
      "language_loss": 0.82024819,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84318721,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 2.7320775985717773
    },
    {
      "auxiliary_loss_clip": 0.01244572,
      "auxiliary_loss_mlp": 0.01047396,
      "balance_loss_clip": 1.03211987,
      "balance_loss_mlp": 1.03332877,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 1.8907299375438693,
      "language_loss": 0.75800443,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.7809242,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.6751673221588135
    },
    {
      "auxiliary_loss_clip": 0.01241317,
      "auxiliary_loss_mlp": 0.01129617,
      "balance_loss_clip": 0.98844409,
      "balance_loss_mlp": 0.0,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.777668073874363,
      "language_loss": 0.73216581,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.75587517,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.6854026317596436
    },
    {
      "auxiliary_loss_clip": 0.01250849,
      "auxiliary_loss_mlp": 0.01051302,
      "balance_loss_clip": 1.02937567,
      "balance_loss_mlp": 1.03718209,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.9444778930758746,
      "language_loss": 0.78541052,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.80843204,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 2.981329917907715
    },
    {
      "auxiliary_loss_clip": 0.0125089,
      "auxiliary_loss_mlp": 0.01057375,
      "balance_loss_clip": 1.03203702,
      "balance_loss_mlp": 1.04262841,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.3983907343223327,
      "language_loss": 0.70730364,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.73038638,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.6387927532196045
    },
    {
      "auxiliary_loss_clip": 0.0124082,
      "auxiliary_loss_mlp": 0.01044695,
      "balance_loss_clip": 0.98717588,
      "balance_loss_mlp": 1.03052115,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 2.9366330969729497,
      "language_loss": 0.91139603,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93425119,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.769258737564087
    },
    {
      "auxiliary_loss_clip": 0.01241352,
      "auxiliary_loss_mlp": 0.01039042,
      "balance_loss_clip": 0.98899722,
      "balance_loss_mlp": 1.02596438,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.9782341043727143,
      "language_loss": 0.82616234,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.84896636,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.815558433532715
    },
    {
      "auxiliary_loss_clip": 0.01237558,
      "auxiliary_loss_mlp": 0.01044233,
      "balance_loss_clip": 0.99010247,
      "balance_loss_mlp": 1.03020215,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.000239439836931,
      "language_loss": 0.81485438,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.83767229,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.6846182346343994
    },
    {
      "auxiliary_loss_clip": 0.01212697,
      "auxiliary_loss_mlp": 0.01054279,
      "balance_loss_clip": 0.90457475,
      "balance_loss_mlp": 1.03999805,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.6213929827555003,
      "language_loss": 0.75214213,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.77481186,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.7253170013427734
    },
    {
      "auxiliary_loss_clip": 0.01243196,
      "auxiliary_loss_mlp": 0.01038055,
      "balance_loss_clip": 0.98965234,
      "balance_loss_mlp": 1.02410758,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 2.1217392741248435,
      "language_loss": 0.85601634,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.87882882,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.6773602962493896
    },
    {
      "auxiliary_loss_clip": 0.01234694,
      "auxiliary_loss_mlp": 0.01041743,
      "balance_loss_clip": 0.98578233,
      "balance_loss_mlp": 1.02740812,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 1.9077555843440834,
      "language_loss": 0.78328282,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80604714,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.7381234169006348
    },
    {
      "auxiliary_loss_clip": 0.01221434,
      "auxiliary_loss_mlp": 0.01043241,
      "balance_loss_clip": 0.94293523,
      "balance_loss_mlp": 1.02947187,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 2.1247780619628407,
      "language_loss": 0.87162375,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89427054,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.7737443447113037
    },
    {
      "auxiliary_loss_clip": 0.01246162,
      "auxiliary_loss_mlp": 0.01041628,
      "balance_loss_clip": 1.06739461,
      "balance_loss_mlp": 1.028157,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 2.3504871591058922,
      "language_loss": 0.82682574,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84970361,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.626084566116333
    },
    {
      "auxiliary_loss_clip": 0.01250487,
      "auxiliary_loss_mlp": 0.01043718,
      "balance_loss_clip": 1.06787455,
      "balance_loss_mlp": 1.02868605,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 3.2904301032425094,
      "language_loss": 0.70149368,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72443569,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.756784439086914
    },
    {
      "auxiliary_loss_clip": 0.01242509,
      "auxiliary_loss_mlp": 0.01048696,
      "balance_loss_clip": 0.95185184,
      "balance_loss_mlp": 1.03449857,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.8491159083453885,
      "language_loss": 0.77997899,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80289102,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.810208320617676
    },
    {
      "auxiliary_loss_clip": 0.01233475,
      "auxiliary_loss_mlp": 0.01043482,
      "balance_loss_clip": 0.9481464,
      "balance_loss_mlp": 1.03013015,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.2118548806690796,
      "language_loss": 0.85432279,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.87709236,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.738759994506836
    },
    {
      "auxiliary_loss_clip": 0.01226704,
      "auxiliary_loss_mlp": 0.01056313,
      "balance_loss_clip": 0.94854486,
      "balance_loss_mlp": 1.04256177,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 2.160805242451491,
      "language_loss": 0.88721597,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.91004616,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.7151246070861816
    },
    {
      "auxiliary_loss_clip": 0.01240344,
      "auxiliary_loss_mlp": 0.01055373,
      "balance_loss_clip": 0.91628516,
      "balance_loss_mlp": 1.04162836,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.6854023017980129,
      "language_loss": 0.82009244,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.84304953,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.7533152103424072
    },
    {
      "auxiliary_loss_clip": 0.01247431,
      "auxiliary_loss_mlp": 0.01045125,
      "balance_loss_clip": 1.03224087,
      "balance_loss_mlp": 1.03051019,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.6851621217444748,
      "language_loss": 0.82304496,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84597051,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 2.715031385421753
    },
    {
      "auxiliary_loss_clip": 0.01240771,
      "auxiliary_loss_mlp": 0.01129836,
      "balance_loss_clip": 1.02993608,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 2.374782846492086,
      "language_loss": 0.79490888,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.81861496,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 2.6155781745910645
    },
    {
      "auxiliary_loss_clip": 0.01242902,
      "auxiliary_loss_mlp": 0.01049927,
      "balance_loss_clip": 1.02922046,
      "balance_loss_mlp": 1.03680205,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 1.8925914285384728,
      "language_loss": 0.84014332,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.86307162,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 2.718285322189331
    },
    {
      "auxiliary_loss_clip": 0.01229594,
      "auxiliary_loss_mlp": 0.01041719,
      "balance_loss_clip": 0.94848877,
      "balance_loss_mlp": 1.02790272,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 1.8519240401712342,
      "language_loss": 0.839113,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.86182618,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 3.6652302742004395
    },
    {
      "auxiliary_loss_clip": 0.01237879,
      "auxiliary_loss_mlp": 0.01044963,
      "balance_loss_clip": 0.98529232,
      "balance_loss_mlp": 1.03154016,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.691738749827971,
      "language_loss": 0.86573702,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.88856548,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 3.788710355758667
    },
    {
      "auxiliary_loss_clip": 0.01244225,
      "auxiliary_loss_mlp": 0.01037991,
      "balance_loss_clip": 1.02582383,
      "balance_loss_mlp": 1.02430558,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 2.44266695959853,
      "language_loss": 0.86191416,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.8847363,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 4.541559934616089
    },
    {
      "auxiliary_loss_clip": 0.01211501,
      "auxiliary_loss_mlp": 0.01057276,
      "balance_loss_clip": 0.866413,
      "balance_loss_mlp": 1.04401994,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 2.271263374993002,
      "language_loss": 0.81655383,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.83924156,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.805371046066284
    },
    {
      "auxiliary_loss_clip": 0.01242831,
      "auxiliary_loss_mlp": 0.01049771,
      "balance_loss_clip": 0.99196184,
      "balance_loss_mlp": 1.03515601,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 2.7204862716706963,
      "language_loss": 0.92216384,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94508982,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 2.6584877967834473
    },
    {
      "auxiliary_loss_clip": 0.01224876,
      "auxiliary_loss_mlp": 0.0104061,
      "balance_loss_clip": 0.90915847,
      "balance_loss_mlp": 1.02660298,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 2.570112077755836,
      "language_loss": 0.85026211,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.87291694,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.773401975631714
    },
    {
      "auxiliary_loss_clip": 0.01227614,
      "auxiliary_loss_mlp": 0.01051026,
      "balance_loss_clip": 0.91007406,
      "balance_loss_mlp": 1.03654242,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 2.088544843392585,
      "language_loss": 0.90683651,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.92962289,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.8227875232696533
    },
    {
      "auxiliary_loss_clip": 0.01229709,
      "auxiliary_loss_mlp": 0.0112944,
      "balance_loss_clip": 0.94608235,
      "balance_loss_mlp": 0.0,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 1.9332878625074885,
      "language_loss": 0.83615005,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.85974157,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.8287193775177
    },
    {
      "auxiliary_loss_clip": 0.01238218,
      "auxiliary_loss_mlp": 0.01044802,
      "balance_loss_clip": 0.99029839,
      "balance_loss_mlp": 1.0307591,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 2.0377148997506134,
      "language_loss": 0.8244521,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.84728229,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.724259853363037
    },
    {
      "auxiliary_loss_clip": 0.01237824,
      "auxiliary_loss_mlp": 0.01043371,
      "balance_loss_clip": 0.99247575,
      "balance_loss_mlp": 1.02882791,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 2.029578220638217,
      "language_loss": 0.84961998,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.87243187,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.7520484924316406
    },
    {
      "auxiliary_loss_clip": 0.01219238,
      "auxiliary_loss_mlp": 0.01049408,
      "balance_loss_clip": 0.90538323,
      "balance_loss_mlp": 1.03543615,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 1.7708553902404338,
      "language_loss": 0.81496716,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.83765358,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.788696765899658
    },
    {
      "auxiliary_loss_clip": 0.0124459,
      "auxiliary_loss_mlp": 0.01037581,
      "balance_loss_clip": 0.94704461,
      "balance_loss_mlp": 1.0236392,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 1.8530300903047137,
      "language_loss": 0.836721,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.85954267,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.802902936935425
    },
    {
      "auxiliary_loss_clip": 0.01251266,
      "auxiliary_loss_mlp": 0.01047276,
      "balance_loss_clip": 1.03188622,
      "balance_loss_mlp": 1.03344798,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 1.902106147249991,
      "language_loss": 0.87738836,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.90037382,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.663503408432007
    },
    {
      "auxiliary_loss_clip": 0.01202308,
      "auxiliary_loss_mlp": 0.01044771,
      "balance_loss_clip": 0.83003187,
      "balance_loss_mlp": 1.03064454,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 3.4166814125052816,
      "language_loss": 0.89414144,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.91661227,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.7544078826904297
    },
    {
      "auxiliary_loss_clip": 0.01215374,
      "auxiliary_loss_mlp": 0.01044524,
      "balance_loss_clip": 0.944148,
      "balance_loss_mlp": 1.03064752,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 1.8725230553798085,
      "language_loss": 0.79077452,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81337351,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.736635446548462
    },
    {
      "auxiliary_loss_clip": 0.01233518,
      "auxiliary_loss_mlp": 0.01040427,
      "balance_loss_clip": 1.02865934,
      "balance_loss_mlp": 1.02630031,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 2.247879383790113,
      "language_loss": 0.86611164,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.88885105,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.608745574951172
    },
    {
      "auxiliary_loss_clip": 0.01245501,
      "auxiliary_loss_mlp": 0.01052002,
      "balance_loss_clip": 1.0302074,
      "balance_loss_mlp": 1.03721964,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 1.9339143128424945,
      "language_loss": 0.84849036,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.87146533,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.65287184715271
    },
    {
      "auxiliary_loss_clip": 0.0124304,
      "auxiliary_loss_mlp": 0.0103997,
      "balance_loss_clip": 1.0674715,
      "balance_loss_mlp": 1.02646363,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.5466473630200472,
      "language_loss": 0.80166042,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82449055,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.637359619140625
    },
    {
      "auxiliary_loss_clip": 0.01234561,
      "auxiliary_loss_mlp": 0.0104526,
      "balance_loss_clip": 0.98764241,
      "balance_loss_mlp": 1.0320636,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 1.7887107271731475,
      "language_loss": 0.64987791,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.67267615,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.7027862071990967
    },
    {
      "auxiliary_loss_clip": 0.01240905,
      "auxiliary_loss_mlp": 0.01044642,
      "balance_loss_clip": 0.95302457,
      "balance_loss_mlp": 1.03142726,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 2.1921487575020486,
      "language_loss": 0.75163805,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.77449358,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.7194364070892334
    },
    {
      "auxiliary_loss_clip": 0.0123119,
      "auxiliary_loss_mlp": 0.01042871,
      "balance_loss_clip": 1.02787733,
      "balance_loss_mlp": 1.02919722,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 3.216062937479027,
      "language_loss": 0.75539923,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.77813983,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.6604483127593994
    },
    {
      "auxiliary_loss_clip": 0.01241755,
      "auxiliary_loss_mlp": 0.01129496,
      "balance_loss_clip": 0.98907208,
      "balance_loss_mlp": 0.0,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 1.7601170045093366,
      "language_loss": 0.7633338,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78704631,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.7480719089508057
    },
    {
      "auxiliary_loss_clip": 0.01230012,
      "auxiliary_loss_mlp": 0.01046779,
      "balance_loss_clip": 0.99063766,
      "balance_loss_mlp": 1.03409481,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 2.084250258788493,
      "language_loss": 0.7961337,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81890154,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 2.777332067489624
    },
    {
      "auxiliary_loss_clip": 0.01236719,
      "auxiliary_loss_mlp": 0.01041878,
      "balance_loss_clip": 0.98723638,
      "balance_loss_mlp": 1.0279429,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 1.8873674281680453,
      "language_loss": 0.75140733,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77419335,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.786381483078003
    },
    {
      "auxiliary_loss_clip": 0.01235064,
      "auxiliary_loss_mlp": 0.0104912,
      "balance_loss_clip": 0.94953668,
      "balance_loss_mlp": 1.03509498,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 1.4958910424832517,
      "language_loss": 0.82604885,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.84889072,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 2.724550485610962
    },
    {
      "auxiliary_loss_clip": 0.01243306,
      "auxiliary_loss_mlp": 0.01043865,
      "balance_loss_clip": 1.06416512,
      "balance_loss_mlp": 1.02994776,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 2.5690663601867496,
      "language_loss": 0.8978014,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92067313,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 2.6463065147399902
    },
    {
      "auxiliary_loss_clip": 0.01229095,
      "auxiliary_loss_mlp": 0.01043754,
      "balance_loss_clip": 0.9886511,
      "balance_loss_mlp": 1.02993727,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 2.025630547716018,
      "language_loss": 0.66223001,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.68495846,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 2.8132784366607666
    },
    {
      "auxiliary_loss_clip": 0.01235531,
      "auxiliary_loss_mlp": 0.01039951,
      "balance_loss_clip": 0.98888469,
      "balance_loss_mlp": 1.02602744,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 2.253686784848359,
      "language_loss": 0.87913752,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.9018923,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 4.6052470207214355
    },
    {
      "auxiliary_loss_clip": 0.01246403,
      "auxiliary_loss_mlp": 0.01041865,
      "balance_loss_clip": 1.03389049,
      "balance_loss_mlp": 1.02885938,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 1.915408123824362,
      "language_loss": 0.72009313,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74297583,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 2.676649332046509
    },
    {
      "auxiliary_loss_clip": 0.01245741,
      "auxiliary_loss_mlp": 0.01040978,
      "balance_loss_clip": 1.067541,
      "balance_loss_mlp": 1.02775753,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 2.079789048810317,
      "language_loss": 0.84096104,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86382824,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 3.746967315673828
    },
    {
      "auxiliary_loss_clip": 0.01240819,
      "auxiliary_loss_mlp": 0.01045134,
      "balance_loss_clip": 0.99225193,
      "balance_loss_mlp": 1.03184175,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 3.071676116865211,
      "language_loss": 0.70702952,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.72988904,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 2.763836145401001
    },
    {
      "auxiliary_loss_clip": 0.01241649,
      "auxiliary_loss_mlp": 0.01042686,
      "balance_loss_clip": 1.03216302,
      "balance_loss_mlp": 1.0293467,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 1.8193195213981121,
      "language_loss": 0.77525008,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.79809344,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.64924693107605
    },
    {
      "auxiliary_loss_clip": 0.01231208,
      "auxiliary_loss_mlp": 0.01045364,
      "balance_loss_clip": 1.0281527,
      "balance_loss_mlp": 1.03197694,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 1.7438408205493736,
      "language_loss": 0.94639611,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96916187,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.722853422164917
    },
    {
      "auxiliary_loss_clip": 0.01220766,
      "auxiliary_loss_mlp": 0.01041498,
      "balance_loss_clip": 0.87003589,
      "balance_loss_mlp": 1.02799153,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 2.904188305733763,
      "language_loss": 0.75211048,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.77473307,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.7853171825408936
    },
    {
      "auxiliary_loss_clip": 0.01236812,
      "auxiliary_loss_mlp": 0.01044097,
      "balance_loss_clip": 0.91391224,
      "balance_loss_mlp": 1.03154421,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 4.02191044160906,
      "language_loss": 0.75877774,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.78158677,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.8100104331970215
    },
    {
      "auxiliary_loss_clip": 0.01227043,
      "auxiliary_loss_mlp": 0.01037458,
      "balance_loss_clip": 0.98934877,
      "balance_loss_mlp": 1.0246129,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 2.334814048202528,
      "language_loss": 0.81683636,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.83948135,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.8697144985198975
    },
    {
      "auxiliary_loss_clip": 0.01239448,
      "auxiliary_loss_mlp": 0.0105478,
      "balance_loss_clip": 0.95510817,
      "balance_loss_mlp": 1.04115438,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 2.0192174150327817,
      "language_loss": 0.88572019,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.90866244,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.863839864730835
    },
    {
      "auxiliary_loss_clip": 0.01241113,
      "auxiliary_loss_mlp": 0.01042619,
      "balance_loss_clip": 0.99280906,
      "balance_loss_mlp": 1.02977967,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 2.5419708850930025,
      "language_loss": 0.77032578,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.79316306,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.718507766723633
    },
    {
      "auxiliary_loss_clip": 0.01227482,
      "auxiliary_loss_mlp": 0.01043664,
      "balance_loss_clip": 0.98964536,
      "balance_loss_mlp": 1.03108132,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.007562757860166,
      "language_loss": 0.88645804,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.90916944,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.7485082149505615
    },
    {
      "auxiliary_loss_clip": 0.01234625,
      "auxiliary_loss_mlp": 0.01041794,
      "balance_loss_clip": 1.02630401,
      "balance_loss_mlp": 1.02834678,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 2.3312135068126767,
      "language_loss": 0.7255367,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.74830091,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.63142991065979
    },
    {
      "auxiliary_loss_clip": 0.01225391,
      "auxiliary_loss_mlp": 0.01038711,
      "balance_loss_clip": 0.91158473,
      "balance_loss_mlp": 1.02588379,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.5928490332850995,
      "language_loss": 0.78455806,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.807199,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.8058924674987793
    },
    {
      "auxiliary_loss_clip": 0.01244592,
      "auxiliary_loss_mlp": 0.01044661,
      "balance_loss_clip": 1.06516409,
      "balance_loss_mlp": 1.03130937,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 2.9384378069757866,
      "language_loss": 0.83059579,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.85348827,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.7506258487701416
    },
    {
      "auxiliary_loss_clip": 0.01225817,
      "auxiliary_loss_mlp": 0.01050558,
      "balance_loss_clip": 0.98607779,
      "balance_loss_mlp": 1.03689635,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 4.371925969018296,
      "language_loss": 0.91772717,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.94049096,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.814249038696289
    },
    {
      "auxiliary_loss_clip": 0.01237219,
      "auxiliary_loss_mlp": 0.01044539,
      "balance_loss_clip": 1.02640724,
      "balance_loss_mlp": 1.03085935,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 1.9769544734927338,
      "language_loss": 0.82852656,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.85134411,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.686048984527588
    },
    {
      "auxiliary_loss_clip": 0.01241092,
      "auxiliary_loss_mlp": 0.01044018,
      "balance_loss_clip": 1.02865791,
      "balance_loss_mlp": 1.03075027,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.3123538187543673,
      "language_loss": 0.70102823,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.72387934,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.9076857566833496
    },
    {
      "auxiliary_loss_clip": 0.01234588,
      "auxiliary_loss_mlp": 0.01129067,
      "balance_loss_clip": 0.94952846,
      "balance_loss_mlp": 0.0,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 1.9455597893065562,
      "language_loss": 0.78974825,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.81338483,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.7463879585266113
    },
    {
      "auxiliary_loss_clip": 0.01239917,
      "auxiliary_loss_mlp": 0.01048232,
      "balance_loss_clip": 0.95276392,
      "balance_loss_mlp": 1.0353992,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 2.1036485359164625,
      "language_loss": 0.90168232,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.92456383,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.717036485671997
    },
    {
      "auxiliary_loss_clip": 0.01241004,
      "auxiliary_loss_mlp": 0.01129089,
      "balance_loss_clip": 0.98797518,
      "balance_loss_mlp": 0.0,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 2.1702323056336916,
      "language_loss": 0.76002872,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78372967,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 2.6608242988586426
    },
    {
      "auxiliary_loss_clip": 0.01239644,
      "auxiliary_loss_mlp": 0.01040855,
      "balance_loss_clip": 0.99227762,
      "balance_loss_mlp": 1.02748013,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.1410884628788533,
      "language_loss": 0.72734976,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.75015479,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.6695621013641357
    },
    {
      "auxiliary_loss_clip": 0.01242741,
      "auxiliary_loss_mlp": 0.01046852,
      "balance_loss_clip": 1.02973866,
      "balance_loss_mlp": 1.03339326,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 3.242848530281079,
      "language_loss": 0.88468969,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90758562,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.613420248031616
    },
    {
      "auxiliary_loss_clip": 0.01162985,
      "auxiliary_loss_mlp": 0.01008095,
      "balance_loss_clip": 0.97622454,
      "balance_loss_mlp": 1.00041795,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9842836020406001,
      "language_loss": 0.64765674,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66936761,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 3.208634614944458
    },
    {
      "auxiliary_loss_clip": 0.01226163,
      "auxiliary_loss_mlp": 0.01039663,
      "balance_loss_clip": 0.98901898,
      "balance_loss_mlp": 1.02618647,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 3.4432989437903063,
      "language_loss": 0.95217228,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.97483051,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 2.7247962951660156
    },
    {
      "auxiliary_loss_clip": 0.01245029,
      "auxiliary_loss_mlp": 0.01039835,
      "balance_loss_clip": 1.06721497,
      "balance_loss_mlp": 1.02608991,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 3.076983470956357,
      "language_loss": 0.66192901,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68477768,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 2.6388988494873047
    },
    {
      "auxiliary_loss_clip": 0.01236809,
      "auxiliary_loss_mlp": 0.01038263,
      "balance_loss_clip": 1.02608323,
      "balance_loss_mlp": 1.02528059,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.4408800973429177,
      "language_loss": 0.79623902,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81898969,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 4.580037593841553
    },
    {
      "auxiliary_loss_clip": 0.01239612,
      "auxiliary_loss_mlp": 0.01040745,
      "balance_loss_clip": 0.91237068,
      "balance_loss_mlp": 1.02794766,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.080717016702495,
      "language_loss": 0.85817128,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.88097489,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 3.7340235710144043
    },
    {
      "auxiliary_loss_clip": 0.01243495,
      "auxiliary_loss_mlp": 0.0104416,
      "balance_loss_clip": 1.03201079,
      "balance_loss_mlp": 1.03139877,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 2.1350181289953136,
      "language_loss": 0.76199573,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78487229,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 3.822282552719116
    },
    {
      "auxiliary_loss_clip": 0.01236682,
      "auxiliary_loss_mlp": 0.01038503,
      "balance_loss_clip": 0.91099209,
      "balance_loss_mlp": 1.02535367,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 2.3403730560868232,
      "language_loss": 0.72507846,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74783033,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 2.729619026184082
    },
    {
      "auxiliary_loss_clip": 0.01238964,
      "auxiliary_loss_mlp": 0.01040613,
      "balance_loss_clip": 0.94640625,
      "balance_loss_mlp": 1.02782238,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 6.30443242250921,
      "language_loss": 0.72008646,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.74288225,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 2.738476514816284
    },
    {
      "auxiliary_loss_clip": 0.01240379,
      "auxiliary_loss_mlp": 0.01047993,
      "balance_loss_clip": 1.02811038,
      "balance_loss_mlp": 1.03546369,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 2.5479312992582996,
      "language_loss": 0.88785428,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91073799,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 2.683885335922241
    },
    {
      "auxiliary_loss_clip": 0.01235008,
      "auxiliary_loss_mlp": 0.01041122,
      "balance_loss_clip": 0.98713797,
      "balance_loss_mlp": 1.02868879,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 2.831666372265466,
      "language_loss": 0.77972019,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80248153,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.728947639465332
    },
    {
      "auxiliary_loss_clip": 0.01149722,
      "auxiliary_loss_mlp": 0.0100671,
      "balance_loss_clip": 0.92453015,
      "balance_loss_mlp": 0.99893761,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 0.9963456413941629,
      "language_loss": 0.67687374,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.69843811,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.240788698196411
    },
    {
      "auxiliary_loss_clip": 0.01216931,
      "auxiliary_loss_mlp": 0.01036176,
      "balance_loss_clip": 0.90633833,
      "balance_loss_mlp": 1.02367687,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 1.659343603569369,
      "language_loss": 0.73622257,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.75875366,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.7614383697509766
    },
    {
      "auxiliary_loss_clip": 0.01234263,
      "auxiliary_loss_mlp": 0.01036835,
      "balance_loss_clip": 1.0252316,
      "balance_loss_mlp": 1.02367473,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.730122048826382,
      "language_loss": 0.88039732,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90310836,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.7158286571502686
    },
    {
      "auxiliary_loss_clip": 0.01242848,
      "auxiliary_loss_mlp": 0.01046562,
      "balance_loss_clip": 1.03129005,
      "balance_loss_mlp": 1.03290057,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 1.6631673650026868,
      "language_loss": 0.78256077,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80545491,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.9484362602233887
    },
    {
      "auxiliary_loss_clip": 0.01243823,
      "auxiliary_loss_mlp": 0.01044903,
      "balance_loss_clip": 0.98773026,
      "balance_loss_mlp": 1.0325408,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 1.7718868416404956,
      "language_loss": 0.79153848,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81442571,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.721900701522827
    },
    {
      "auxiliary_loss_clip": 0.0114564,
      "auxiliary_loss_mlp": 0.01124674,
      "balance_loss_clip": 0.96637285,
      "balance_loss_mlp": 0.0,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8186341488115472,
      "language_loss": 0.56737781,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.59008092,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.271026611328125
    },
    {
      "auxiliary_loss_clip": 0.01240335,
      "auxiliary_loss_mlp": 0.01044955,
      "balance_loss_clip": 0.99288934,
      "balance_loss_mlp": 1.03309393,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 1.7631214802983586,
      "language_loss": 0.82124233,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84409529,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.7575602531433105
    },
    {
      "auxiliary_loss_clip": 0.01235763,
      "auxiliary_loss_mlp": 0.01128959,
      "balance_loss_clip": 0.99434054,
      "balance_loss_mlp": 0.0,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.7509929041651418,
      "language_loss": 0.8084172,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.83206439,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.778214931488037
    },
    {
      "auxiliary_loss_clip": 0.01231659,
      "auxiliary_loss_mlp": 0.01043119,
      "balance_loss_clip": 0.91350579,
      "balance_loss_mlp": 1.02960062,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 3.2815265481953597,
      "language_loss": 0.87454808,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.89729583,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.8459107875823975
    },
    {
      "auxiliary_loss_clip": 0.0123721,
      "auxiliary_loss_mlp": 0.01041676,
      "balance_loss_clip": 0.98755252,
      "balance_loss_mlp": 1.02847958,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.524808117789222,
      "language_loss": 0.86596131,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.8887502,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.67317795753479
    },
    {
      "auxiliary_loss_clip": 0.01239843,
      "auxiliary_loss_mlp": 0.0104624,
      "balance_loss_clip": 0.94875801,
      "balance_loss_mlp": 1.03320456,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 2.252665352894258,
      "language_loss": 0.86157465,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.88443547,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.6492562294006348
    },
    {
      "auxiliary_loss_clip": 0.01242706,
      "auxiliary_loss_mlp": 0.01043539,
      "balance_loss_clip": 1.02767873,
      "balance_loss_mlp": 1.03084922,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 1.8743385221079862,
      "language_loss": 0.69456816,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71743059,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.68941593170166
    },
    {
      "auxiliary_loss_clip": 0.01236584,
      "auxiliary_loss_mlp": 0.01039097,
      "balance_loss_clip": 0.98846954,
      "balance_loss_mlp": 1.02612734,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 2.059265473419083,
      "language_loss": 0.79155707,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81431401,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.677626132965088
    },
    {
      "auxiliary_loss_clip": 0.01241012,
      "auxiliary_loss_mlp": 0.01039827,
      "balance_loss_clip": 1.06715572,
      "balance_loss_mlp": 1.02693403,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 2.0407573997979065,
      "language_loss": 0.72384018,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74664855,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.611727237701416
    },
    {
      "auxiliary_loss_clip": 0.01233845,
      "auxiliary_loss_mlp": 0.01040876,
      "balance_loss_clip": 0.91168582,
      "balance_loss_mlp": 1.0287044,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 1.9892294485895519,
      "language_loss": 0.82213616,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.84488332,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 2.7580325603485107
    },
    {
      "auxiliary_loss_clip": 0.01243089,
      "auxiliary_loss_mlp": 0.0104205,
      "balance_loss_clip": 0.99329424,
      "balance_loss_mlp": 1.02928257,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 1.7883153852368865,
      "language_loss": 0.8733958,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89624715,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.6688504219055176
    },
    {
      "auxiliary_loss_clip": 0.01238242,
      "auxiliary_loss_mlp": 0.01042098,
      "balance_loss_clip": 1.02743697,
      "balance_loss_mlp": 1.02895474,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 1.7853712805205773,
      "language_loss": 0.80647761,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.82928103,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 2.5846118927001953
    },
    {
      "auxiliary_loss_clip": 0.01202624,
      "auxiliary_loss_mlp": 0.0103644,
      "balance_loss_clip": 0.9037472,
      "balance_loss_mlp": 1.02310085,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 1.7747958560397403,
      "language_loss": 0.825661,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.84805167,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 2.766573429107666
    },
    {
      "auxiliary_loss_clip": 0.01223397,
      "auxiliary_loss_mlp": 0.01045045,
      "balance_loss_clip": 0.87154096,
      "balance_loss_mlp": 1.0323019,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 1.9235830519520554,
      "language_loss": 0.87323415,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.89591861,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 3.843247890472412
    },
    {
      "auxiliary_loss_clip": 0.01208849,
      "auxiliary_loss_mlp": 0.01055798,
      "balance_loss_clip": 0.82487309,
      "balance_loss_mlp": 1.04293585,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.227744904467663,
      "language_loss": 0.9762125,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99885899,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 3.8314459323883057
    },
    {
      "auxiliary_loss_clip": 0.01237219,
      "auxiliary_loss_mlp": 0.01042559,
      "balance_loss_clip": 1.02896595,
      "balance_loss_mlp": 1.02988744,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 2.410747087983905,
      "language_loss": 0.92307496,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94587272,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 3.9229350090026855
    },
    {
      "auxiliary_loss_clip": 0.0123245,
      "auxiliary_loss_mlp": 0.01041952,
      "balance_loss_clip": 0.94588637,
      "balance_loss_mlp": 1.02889252,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 1.8982372668274068,
      "language_loss": 0.736395,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.75913894,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 3.8685195446014404
    },
    {
      "auxiliary_loss_clip": 0.01241484,
      "auxiliary_loss_mlp": 0.0103681,
      "balance_loss_clip": 1.02855051,
      "balance_loss_mlp": 1.02343488,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 1.9456567789252956,
      "language_loss": 0.91794837,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.94073129,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.6029372215270996
    },
    {
      "auxiliary_loss_clip": 0.0114248,
      "auxiliary_loss_mlp": 0.01015408,
      "balance_loss_clip": 0.9238404,
      "balance_loss_mlp": 1.00782609,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.8624428647677352,
      "language_loss": 0.62808549,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.6496644,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.3202598094940186
    },
    {
      "auxiliary_loss_clip": 0.01217335,
      "auxiliary_loss_mlp": 0.01052041,
      "balance_loss_clip": 0.9115721,
      "balance_loss_mlp": 1.03972065,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 3.568449878222541,
      "language_loss": 0.81776762,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.84046131,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 2.759296417236328
    },
    {
      "auxiliary_loss_clip": 0.01243161,
      "auxiliary_loss_mlp": 0.01049942,
      "balance_loss_clip": 1.03217995,
      "balance_loss_mlp": 1.03742492,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 2.2040699140987052,
      "language_loss": 0.75904268,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.78197372,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.709343671798706
    },
    {
      "auxiliary_loss_clip": 0.01237367,
      "auxiliary_loss_mlp": 0.01046477,
      "balance_loss_clip": 1.02765512,
      "balance_loss_mlp": 1.03272676,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 2.1714816148022327,
      "language_loss": 0.80733591,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.83017433,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.7563669681549072
    },
    {
      "auxiliary_loss_clip": 0.01244476,
      "auxiliary_loss_mlp": 0.01129053,
      "balance_loss_clip": 1.06830657,
      "balance_loss_mlp": 0.0,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 3.0656549591242235,
      "language_loss": 0.7849344,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80866969,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.6335866451263428
    },
    {
      "auxiliary_loss_clip": 0.01223218,
      "auxiliary_loss_mlp": 0.01044218,
      "balance_loss_clip": 0.94514167,
      "balance_loss_mlp": 1.03152847,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 2.920447038164132,
      "language_loss": 0.86103302,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.8837074,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.8093018531799316
    },
    {
      "auxiliary_loss_clip": 0.01239881,
      "auxiliary_loss_mlp": 0.01039102,
      "balance_loss_clip": 1.06523538,
      "balance_loss_mlp": 1.02706194,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 2.306795425705535,
      "language_loss": 0.84547007,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.86825991,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.6313507556915283
    },
    {
      "auxiliary_loss_clip": 0.01239853,
      "auxiliary_loss_mlp": 0.01048145,
      "balance_loss_clip": 1.06706131,
      "balance_loss_mlp": 1.03533626,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 1.8581045650611763,
      "language_loss": 0.79311049,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81599039,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.6323068141937256
    },
    {
      "auxiliary_loss_clip": 0.01238903,
      "auxiliary_loss_mlp": 0.01040328,
      "balance_loss_clip": 1.02699625,
      "balance_loss_mlp": 1.02729225,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 2.4852539452868125,
      "language_loss": 0.78367275,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80646503,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.640881299972534
    },
    {
      "auxiliary_loss_clip": 0.01235974,
      "auxiliary_loss_mlp": 0.01050967,
      "balance_loss_clip": 1.06257129,
      "balance_loss_mlp": 1.03840792,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 1.9165484921538805,
      "language_loss": 0.78064269,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.8035121,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.7487714290618896
    },
    {
      "auxiliary_loss_clip": 0.01238326,
      "auxiliary_loss_mlp": 0.01129428,
      "balance_loss_clip": 0.98809415,
      "balance_loss_mlp": 0.0,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 1.8720631823244405,
      "language_loss": 0.81624597,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.8399235,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.7274484634399414
    },
    {
      "auxiliary_loss_clip": 0.01230815,
      "auxiliary_loss_mlp": 0.01049219,
      "balance_loss_clip": 0.95254683,
      "balance_loss_mlp": 1.03707778,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 1.810198069234511,
      "language_loss": 0.75148392,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77428424,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.728729724884033
    },
    {
      "auxiliary_loss_clip": 0.01245313,
      "auxiliary_loss_mlp": 0.0104231,
      "balance_loss_clip": 1.02667236,
      "balance_loss_mlp": 1.02950668,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 2.3998734893851785,
      "language_loss": 0.84802568,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.87090194,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.698988437652588
    },
    {
      "auxiliary_loss_clip": 0.01238176,
      "auxiliary_loss_mlp": 0.01042044,
      "balance_loss_clip": 0.98550326,
      "balance_loss_mlp": 1.02965808,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 2.460169788009642,
      "language_loss": 0.55713809,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.57994032,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.747149705886841
    },
    {
      "auxiliary_loss_clip": 0.01237635,
      "auxiliary_loss_mlp": 0.01043142,
      "balance_loss_clip": 1.06578732,
      "balance_loss_mlp": 1.02988577,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 1.7067586089133564,
      "language_loss": 0.87465632,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.8974641,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.773541212081909
    },
    {
      "auxiliary_loss_clip": 0.01226578,
      "auxiliary_loss_mlp": 0.01038378,
      "balance_loss_clip": 0.9467274,
      "balance_loss_mlp": 1.02591515,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 1.7679171845384511,
      "language_loss": 0.82475525,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.84740478,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.8017654418945312
    },
    {
      "auxiliary_loss_clip": 0.01241085,
      "auxiliary_loss_mlp": 0.01040314,
      "balance_loss_clip": 1.02933466,
      "balance_loss_mlp": 1.02700424,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 1.8099660569893408,
      "language_loss": 0.91274482,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.9355588,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.6250412464141846
    },
    {
      "auxiliary_loss_clip": 0.01238716,
      "auxiliary_loss_mlp": 0.0103666,
      "balance_loss_clip": 1.06418657,
      "balance_loss_mlp": 1.02365422,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.7295663528219876,
      "language_loss": 0.84009612,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86284983,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.614899158477783
    },
    {
      "auxiliary_loss_clip": 0.012349,
      "auxiliary_loss_mlp": 0.0103979,
      "balance_loss_clip": 0.9472391,
      "balance_loss_mlp": 1.02666545,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 1.937529965841755,
      "language_loss": 0.84087968,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.8636266,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 2.8100788593292236
    },
    {
      "auxiliary_loss_clip": 0.01237558,
      "auxiliary_loss_mlp": 0.01036945,
      "balance_loss_clip": 0.98998106,
      "balance_loss_mlp": 1.02466655,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 1.9788970179480048,
      "language_loss": 0.84148365,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86422873,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.6426479816436768
    },
    {
      "auxiliary_loss_clip": 0.01230724,
      "auxiliary_loss_mlp": 0.01044463,
      "balance_loss_clip": 1.02348804,
      "balance_loss_mlp": 1.03209496,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 2.6522484504328365,
      "language_loss": 0.69331169,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.7160635,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 2.721519947052002
    },
    {
      "auxiliary_loss_clip": 0.01141782,
      "auxiliary_loss_mlp": 0.0101384,
      "balance_loss_clip": 0.97022963,
      "balance_loss_mlp": 1.00640142,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.8256668241946233,
      "language_loss": 0.60886747,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.63042367,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 4.188354730606079
    },
    {
      "auxiliary_loss_clip": 0.01218434,
      "auxiliary_loss_mlp": 0.01040716,
      "balance_loss_clip": 0.98336804,
      "balance_loss_mlp": 1.02815771,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 2.41305271897889,
      "language_loss": 0.74728537,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.76987684,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 3.7952871322631836
    },
    {
      "auxiliary_loss_clip": 0.01242884,
      "auxiliary_loss_mlp": 0.01047278,
      "balance_loss_clip": 1.0683161,
      "balance_loss_mlp": 1.0348264,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 3.8403859658660995,
      "language_loss": 0.85424626,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87714791,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 3.5809082984924316
    },
    {
      "auxiliary_loss_clip": 0.01230534,
      "auxiliary_loss_mlp": 0.0104399,
      "balance_loss_clip": 0.98564446,
      "balance_loss_mlp": 1.03110385,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 2.7740046681443595,
      "language_loss": 0.74260819,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.76535344,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 3.658972978591919
    },
    {
      "auxiliary_loss_clip": 0.01235665,
      "auxiliary_loss_mlp": 0.01047223,
      "balance_loss_clip": 1.06704926,
      "balance_loss_mlp": 1.03485465,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.260634504839598,
      "language_loss": 0.89699614,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91982502,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 2.5883326530456543
    },
    {
      "auxiliary_loss_clip": 0.01237515,
      "auxiliary_loss_mlp": 0.01043835,
      "balance_loss_clip": 1.02930856,
      "balance_loss_mlp": 1.0308888,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 3.9090456434322047,
      "language_loss": 0.81727129,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.84008479,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.6358118057250977
    },
    {
      "auxiliary_loss_clip": 0.01236875,
      "auxiliary_loss_mlp": 0.01034948,
      "balance_loss_clip": 1.02590179,
      "balance_loss_mlp": 1.02275276,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 2.1266797678398364,
      "language_loss": 0.81629324,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.83901143,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.695277214050293
    },
    {
      "auxiliary_loss_clip": 0.01238358,
      "auxiliary_loss_mlp": 0.01045781,
      "balance_loss_clip": 1.0643301,
      "balance_loss_mlp": 1.03324616,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.902755654114165,
      "language_loss": 0.81277192,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83561325,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 2.603999137878418
    },
    {
      "auxiliary_loss_clip": 0.01142145,
      "auxiliary_loss_mlp": 0.01008905,
      "balance_loss_clip": 0.96879363,
      "balance_loss_mlp": 1.00175273,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.706943319684227,
      "language_loss": 0.52634895,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54785943,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.2305378913879395
    },
    {
      "auxiliary_loss_clip": 0.01222932,
      "auxiliary_loss_mlp": 0.01043402,
      "balance_loss_clip": 0.94748783,
      "balance_loss_mlp": 1.03081954,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 2.1243423476693977,
      "language_loss": 0.8122133,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.8348766,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.7537877559661865
    },
    {
      "auxiliary_loss_clip": 0.0123327,
      "auxiliary_loss_mlp": 0.01035245,
      "balance_loss_clip": 0.94961643,
      "balance_loss_mlp": 1.02296615,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.189155303241346,
      "language_loss": 0.79712045,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.81980556,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.7639639377593994
    },
    {
      "auxiliary_loss_clip": 0.01222137,
      "auxiliary_loss_mlp": 0.0104406,
      "balance_loss_clip": 0.94745016,
      "balance_loss_mlp": 1.03151321,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 2.0204685773964526,
      "language_loss": 0.83195031,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85461229,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.8116071224212646
    },
    {
      "auxiliary_loss_clip": 0.012177,
      "auxiliary_loss_mlp": 0.01038611,
      "balance_loss_clip": 0.82592839,
      "balance_loss_mlp": 1.02602255,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 1.6475321308741258,
      "language_loss": 0.88062739,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90319049,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.9015517234802246
    },
    {
      "auxiliary_loss_clip": 0.01224796,
      "auxiliary_loss_mlp": 0.0105219,
      "balance_loss_clip": 0.9448545,
      "balance_loss_mlp": 1.03906476,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 2.1572424709768523,
      "language_loss": 0.76381862,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.78658855,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.8142619132995605
    },
    {
      "auxiliary_loss_clip": 0.0123264,
      "auxiliary_loss_mlp": 0.01041634,
      "balance_loss_clip": 0.98689437,
      "balance_loss_mlp": 1.0294745,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 1.8629005635263078,
      "language_loss": 0.82806885,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.8508116,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.760676860809326
    },
    {
      "auxiliary_loss_clip": 0.01234608,
      "auxiliary_loss_mlp": 0.010429,
      "balance_loss_clip": 1.02545178,
      "balance_loss_mlp": 1.03104496,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.7144472279186287,
      "language_loss": 0.83139539,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85417044,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.753506898880005
    },
    {
      "auxiliary_loss_clip": 0.01227559,
      "auxiliary_loss_mlp": 0.01044918,
      "balance_loss_clip": 0.98457909,
      "balance_loss_mlp": 1.03294325,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.4130986825221017,
      "language_loss": 0.78197235,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.80469716,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.7325541973114014
    },
    {
      "auxiliary_loss_clip": 0.01220985,
      "auxiliary_loss_mlp": 0.01043093,
      "balance_loss_clip": 1.02422035,
      "balance_loss_mlp": 1.03093362,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 1.6829051855992012,
      "language_loss": 0.81054676,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.83318758,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.7737293243408203
    },
    {
      "auxiliary_loss_clip": 0.0124358,
      "auxiliary_loss_mlp": 0.01047837,
      "balance_loss_clip": 0.9867382,
      "balance_loss_mlp": 1.03502202,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 1.776435340740662,
      "language_loss": 0.74897069,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77188492,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.764070987701416
    },
    {
      "auxiliary_loss_clip": 0.0120982,
      "auxiliary_loss_mlp": 0.01039951,
      "balance_loss_clip": 0.90445697,
      "balance_loss_mlp": 1.02740407,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.416846899509836,
      "language_loss": 0.86337662,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88587439,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.8652729988098145
    },
    {
      "auxiliary_loss_clip": 0.01215202,
      "auxiliary_loss_mlp": 0.01040358,
      "balance_loss_clip": 0.86562669,
      "balance_loss_mlp": 1.02757835,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 5.351164917160013,
      "language_loss": 0.77325135,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.79580694,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.809500217437744
    },
    {
      "auxiliary_loss_clip": 0.01227239,
      "auxiliary_loss_mlp": 0.01047822,
      "balance_loss_clip": 0.90941364,
      "balance_loss_mlp": 1.03628826,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 2.2776327135883587,
      "language_loss": 0.80834156,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.83109224,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.8260974884033203
    },
    {
      "auxiliary_loss_clip": 0.01239222,
      "auxiliary_loss_mlp": 0.01040398,
      "balance_loss_clip": 1.06682563,
      "balance_loss_mlp": 1.02883494,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 1.797488217091338,
      "language_loss": 0.7936188,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81641495,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 2.699359655380249
    },
    {
      "auxiliary_loss_clip": 0.01232852,
      "auxiliary_loss_mlp": 0.01040266,
      "balance_loss_clip": 1.02788007,
      "balance_loss_mlp": 1.02743864,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 2.212966424389325,
      "language_loss": 0.88144636,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.90417755,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.679471015930176
    },
    {
      "auxiliary_loss_clip": 0.0123168,
      "auxiliary_loss_mlp": 0.01129366,
      "balance_loss_clip": 0.98698401,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 3.5926277065491505,
      "language_loss": 0.7527225,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.77633297,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 2.837441921234131
    },
    {
      "auxiliary_loss_clip": 0.01238589,
      "auxiliary_loss_mlp": 0.01035083,
      "balance_loss_clip": 0.98434806,
      "balance_loss_mlp": 1.02272153,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 7.428883518641305,
      "language_loss": 0.7947427,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.81747949,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.7675540447235107
    },
    {
      "auxiliary_loss_clip": 0.01227652,
      "auxiliary_loss_mlp": 0.01043135,
      "balance_loss_clip": 1.02465391,
      "balance_loss_mlp": 1.03088641,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 2.994479245077126,
      "language_loss": 0.9144789,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.93718678,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 4.3083906173706055
    },
    {
      "auxiliary_loss_clip": 0.01135597,
      "auxiliary_loss_mlp": 0.01040555,
      "balance_loss_clip": 1.0019424,
      "balance_loss_mlp": 1.03402185,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7865893224661836,
      "language_loss": 0.60915816,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.63091969,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 4.148955821990967
    },
    {
      "auxiliary_loss_clip": 0.01198202,
      "auxiliary_loss_mlp": 0.01036842,
      "balance_loss_clip": 0.82551152,
      "balance_loss_mlp": 1.02516508,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 2.3427094795877013,
      "language_loss": 0.72594452,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.74829495,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 1.1518328189849854
    },
    {
      "auxiliary_loss_clip": 0.01228451,
      "auxiliary_loss_mlp": 0.01041286,
      "balance_loss_clip": 0.94789469,
      "balance_loss_mlp": 1.02866197,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 1.8323694264977026,
      "language_loss": 0.8879016,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91059899,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 3.729907751083374
    },
    {
      "auxiliary_loss_clip": 0.01235403,
      "auxiliary_loss_mlp": 0.01040717,
      "balance_loss_clip": 0.90953076,
      "balance_loss_mlp": 1.02747893,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 2.016285417194981,
      "language_loss": 0.71923339,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74199456,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 2.836951971054077
    },
    {
      "auxiliary_loss_clip": 0.01226124,
      "auxiliary_loss_mlp": 0.01052432,
      "balance_loss_clip": 0.9100132,
      "balance_loss_mlp": 1.04005194,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 2.1073758907978144,
      "language_loss": 0.77019691,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79298252,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.8140432834625244
    },
    {
      "auxiliary_loss_clip": 0.01223301,
      "auxiliary_loss_mlp": 0.01036372,
      "balance_loss_clip": 0.90284288,
      "balance_loss_mlp": 1.02353358,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 2.2042873621197168,
      "language_loss": 0.7746321,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.79722881,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.8294126987457275
    },
    {
      "auxiliary_loss_clip": 0.01231028,
      "auxiliary_loss_mlp": 0.01038047,
      "balance_loss_clip": 0.95134342,
      "balance_loss_mlp": 1.02508903,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 2.2063677222951688,
      "language_loss": 0.71144462,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.73413539,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 2.8399102687835693
    },
    {
      "auxiliary_loss_clip": 0.01138758,
      "auxiliary_loss_mlp": 0.0100962,
      "balance_loss_clip": 0.96159542,
      "balance_loss_mlp": 1.00294387,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.9150952583582795,
      "language_loss": 0.59048074,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61196452,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.4045591354370117
    },
    {
      "auxiliary_loss_clip": 0.01230942,
      "auxiliary_loss_mlp": 0.01040526,
      "balance_loss_clip": 0.9071849,
      "balance_loss_mlp": 1.02878404,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.279997757516046,
      "language_loss": 0.81360036,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83631504,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 3.028249502182007
    },
    {
      "auxiliary_loss_clip": 0.01139562,
      "auxiliary_loss_mlp": 0.010063,
      "balance_loss_clip": 0.99987721,
      "balance_loss_mlp": 0.99952859,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9477252399004898,
      "language_loss": 0.60736108,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.6288197,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.2353291511535645
    },
    {
      "auxiliary_loss_clip": 0.01222023,
      "auxiliary_loss_mlp": 0.01041781,
      "balance_loss_clip": 0.94669878,
      "balance_loss_mlp": 1.02946639,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 2.386611523400928,
      "language_loss": 0.76951766,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79215574,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.990349769592285
    },
    {
      "auxiliary_loss_clip": 0.01233539,
      "auxiliary_loss_mlp": 0.01128894,
      "balance_loss_clip": 0.98811316,
      "balance_loss_mlp": 0.0,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 2.0179991062469433,
      "language_loss": 0.81974423,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.84336853,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.752824306488037
    },
    {
      "auxiliary_loss_clip": 0.01140057,
      "auxiliary_loss_mlp": 0.01009496,
      "balance_loss_clip": 0.85503751,
      "balance_loss_mlp": 1.00248623,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.9677166298691136,
      "language_loss": 0.59396881,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61546433,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.1946420669555664
    },
    {
      "auxiliary_loss_clip": 0.01233782,
      "auxiliary_loss_mlp": 0.01035651,
      "balance_loss_clip": 0.94748431,
      "balance_loss_mlp": 1.0234741,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.582294362076957,
      "language_loss": 0.72324443,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.74593872,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.8221275806427
    },
    {
      "auxiliary_loss_clip": 0.0123381,
      "auxiliary_loss_mlp": 0.0104095,
      "balance_loss_clip": 0.94467944,
      "balance_loss_mlp": 1.0279026,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 2.176648106329632,
      "language_loss": 0.68860376,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.71135139,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.8683626651763916
    },
    {
      "auxiliary_loss_clip": 0.01233113,
      "auxiliary_loss_mlp": 0.01040405,
      "balance_loss_clip": 1.06380224,
      "balance_loss_mlp": 1.02861524,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 1.8752042512698457,
      "language_loss": 0.67836231,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.70109749,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.785754680633545
    },
    {
      "auxiliary_loss_clip": 0.01228828,
      "auxiliary_loss_mlp": 0.01037717,
      "balance_loss_clip": 1.02611041,
      "balance_loss_mlp": 1.02547979,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 1.792672890638599,
      "language_loss": 0.84553164,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.86819708,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.786247968673706
    },
    {
      "auxiliary_loss_clip": 0.01222991,
      "auxiliary_loss_mlp": 0.01045409,
      "balance_loss_clip": 0.98334205,
      "balance_loss_mlp": 1.0333097,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 1.907150597922227,
      "language_loss": 0.87793082,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.90061486,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.9589502811431885
    },
    {
      "auxiliary_loss_clip": 0.01230991,
      "auxiliary_loss_mlp": 0.01045166,
      "balance_loss_clip": 1.02811956,
      "balance_loss_mlp": 1.03310752,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 3.593861964783029,
      "language_loss": 0.69697952,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71974111,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.7944653034210205
    },
    {
      "auxiliary_loss_clip": 0.01206805,
      "auxiliary_loss_mlp": 0.01039652,
      "balance_loss_clip": 0.90309972,
      "balance_loss_mlp": 1.02662849,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 2.4258539449994907,
      "language_loss": 0.92503041,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94749498,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.7123348712921143
    },
    {
      "auxiliary_loss_clip": 0.01213971,
      "auxiliary_loss_mlp": 0.01032383,
      "balance_loss_clip": 0.97962803,
      "balance_loss_mlp": 1.02074802,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 2.0380301002046015,
      "language_loss": 0.75481927,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77728283,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.727754831314087
    },
    {
      "auxiliary_loss_clip": 0.01233487,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.06293082,
      "balance_loss_mlp": 1.02536798,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.6216238509433838,
      "language_loss": 0.78597879,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80868709,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 2.6100823879241943
    },
    {
      "auxiliary_loss_clip": 0.01239191,
      "auxiliary_loss_mlp": 0.01041825,
      "balance_loss_clip": 1.06265187,
      "balance_loss_mlp": 1.02924824,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 2.1437321889612435,
      "language_loss": 0.84705377,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.86986399,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.7019705772399902
    },
    {
      "auxiliary_loss_clip": 0.01224313,
      "auxiliary_loss_mlp": 0.01128928,
      "balance_loss_clip": 0.98514199,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 2.3505019453258504,
      "language_loss": 0.90951514,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.93304753,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 2.7104086875915527
    },
    {
      "auxiliary_loss_clip": 0.0122968,
      "auxiliary_loss_mlp": 0.01040898,
      "balance_loss_clip": 0.98960221,
      "balance_loss_mlp": 1.0287919,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 2.099979396121369,
      "language_loss": 0.90407121,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.92677701,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 2.6922497749328613
    },
    {
      "auxiliary_loss_clip": 0.01141145,
      "auxiliary_loss_mlp": 0.01012363,
      "balance_loss_clip": 0.96166813,
      "balance_loss_mlp": 1.00563931,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 1.0819822501750942,
      "language_loss": 0.56593758,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58747268,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 3.3504269123077393
    },
    {
      "auxiliary_loss_clip": 0.01232501,
      "auxiliary_loss_mlp": 0.01045692,
      "balance_loss_clip": 1.02824759,
      "balance_loss_mlp": 1.03356254,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 1.7313738176673839,
      "language_loss": 0.78415167,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80693364,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 5.316360712051392
    },
    {
      "auxiliary_loss_clip": 0.0123668,
      "auxiliary_loss_mlp": 0.01033521,
      "balance_loss_clip": 1.06662083,
      "balance_loss_mlp": 1.02160573,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 9.066211207573609,
      "language_loss": 0.81769645,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.84039843,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 3.5524415969848633
    },
    {
      "auxiliary_loss_clip": 0.01227594,
      "auxiliary_loss_mlp": 0.01043008,
      "balance_loss_clip": 0.98818803,
      "balance_loss_mlp": 1.03102112,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 1.8438133863351176,
      "language_loss": 0.84480363,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.8675096,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 3.6761999130249023
    },
    {
      "auxiliary_loss_clip": 0.0123942,
      "auxiliary_loss_mlp": 0.01041992,
      "balance_loss_clip": 1.0679009,
      "balance_loss_mlp": 1.02945113,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 2.629205780929822,
      "language_loss": 0.73921424,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.76202834,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 2.705805540084839
    },
    {
      "auxiliary_loss_clip": 0.01135301,
      "auxiliary_loss_mlp": 0.01008355,
      "balance_loss_clip": 0.92454171,
      "balance_loss_mlp": 1.0017271,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8372801753606768,
      "language_loss": 0.54648054,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56791711,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.2912580966949463
    },
    {
      "auxiliary_loss_clip": 0.01237262,
      "auxiliary_loss_mlp": 0.01040572,
      "balance_loss_clip": 1.02855134,
      "balance_loss_mlp": 1.02870512,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 2.1857032776183654,
      "language_loss": 0.70267874,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.72545707,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.6662750244140625
    },
    {
      "auxiliary_loss_clip": 0.01135951,
      "auxiliary_loss_mlp": 0.01010573,
      "balance_loss_clip": 0.92216057,
      "balance_loss_mlp": 1.00370622,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7626443364478569,
      "language_loss": 0.58124143,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60270667,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 3.232752561569214
    },
    {
      "auxiliary_loss_clip": 0.01240204,
      "auxiliary_loss_mlp": 0.01128705,
      "balance_loss_clip": 1.02898049,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 2.1641047479406934,
      "language_loss": 0.73774004,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.76142913,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 2.9111435413360596
    },
    {
      "auxiliary_loss_clip": 0.01212618,
      "auxiliary_loss_mlp": 0.01042419,
      "balance_loss_clip": 0.86919874,
      "balance_loss_mlp": 1.02986681,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 1.6987609650665336,
      "language_loss": 0.8920275,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.9145779,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.8786089420318604
    },
    {
      "auxiliary_loss_clip": 0.01221392,
      "auxiliary_loss_mlp": 0.0104176,
      "balance_loss_clip": 0.91059953,
      "balance_loss_mlp": 1.02977383,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 1.9792098742619173,
      "language_loss": 0.82329893,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84593046,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.7635750770568848
    },
    {
      "auxiliary_loss_clip": 0.01231596,
      "auxiliary_loss_mlp": 0.01037304,
      "balance_loss_clip": 1.02670693,
      "balance_loss_mlp": 1.02599716,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 1.6725308313887117,
      "language_loss": 0.92645931,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94914836,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.747213840484619
    },
    {
      "auxiliary_loss_clip": 0.01212115,
      "auxiliary_loss_mlp": 0.01046171,
      "balance_loss_clip": 0.8327145,
      "balance_loss_mlp": 1.0337733,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 2.0502466159368664,
      "language_loss": 0.78521764,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.80780053,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 3.086962938308716
    },
    {
      "auxiliary_loss_clip": 0.01234059,
      "auxiliary_loss_mlp": 0.01038828,
      "balance_loss_clip": 1.06365037,
      "balance_loss_mlp": 1.02663875,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 1.7591936524882847,
      "language_loss": 0.74504739,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.76777625,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.872965097427368
    },
    {
      "auxiliary_loss_clip": 0.01222835,
      "auxiliary_loss_mlp": 0.01043037,
      "balance_loss_clip": 0.94996512,
      "balance_loss_mlp": 1.03088951,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 1.9074088066836445,
      "language_loss": 0.85209441,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87475312,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.7535109519958496
    },
    {
      "auxiliary_loss_clip": 0.01222613,
      "auxiliary_loss_mlp": 0.01128845,
      "balance_loss_clip": 1.02603614,
      "balance_loss_mlp": 0.0,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 2.406015427565103,
      "language_loss": 0.91383803,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.9373526,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.8265395164489746
    },
    {
      "auxiliary_loss_clip": 0.01238995,
      "auxiliary_loss_mlp": 0.01038963,
      "balance_loss_clip": 1.06697464,
      "balance_loss_mlp": 1.02629709,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.8355830628493353,
      "language_loss": 0.66730797,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.69008756,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.658266067504883
    },
    {
      "auxiliary_loss_clip": 0.01228766,
      "auxiliary_loss_mlp": 0.0103958,
      "balance_loss_clip": 0.9902212,
      "balance_loss_mlp": 1.0271883,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 2.2351881287801874,
      "language_loss": 0.85039473,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87307811,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.7343695163726807
    },
    {
      "auxiliary_loss_clip": 0.01234885,
      "auxiliary_loss_mlp": 0.01037683,
      "balance_loss_clip": 1.02615857,
      "balance_loss_mlp": 1.02638817,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 1.807517175633649,
      "language_loss": 0.74473387,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.76745951,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.9391303062438965
    },
    {
      "auxiliary_loss_clip": 0.0123878,
      "auxiliary_loss_mlp": 0.01042037,
      "balance_loss_clip": 1.02792478,
      "balance_loss_mlp": 1.02971041,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 2.9702016498601425,
      "language_loss": 0.79028523,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.81309342,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.721463203430176
    },
    {
      "auxiliary_loss_clip": 0.01214932,
      "auxiliary_loss_mlp": 0.01128594,
      "balance_loss_clip": 0.98404956,
      "balance_loss_mlp": 0.0,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 4.932944080369545,
      "language_loss": 0.85744429,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.88087964,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.8485989570617676
    },
    {
      "auxiliary_loss_clip": 0.01223962,
      "auxiliary_loss_mlp": 0.01046842,
      "balance_loss_clip": 0.94622898,
      "balance_loss_mlp": 1.03435493,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 3.2899439410363227,
      "language_loss": 0.85825199,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.88095999,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.7564826011657715
    },
    {
      "auxiliary_loss_clip": 0.0121971,
      "auxiliary_loss_mlp": 0.01039215,
      "balance_loss_clip": 0.98505473,
      "balance_loss_mlp": 1.02702558,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 1.6156221221877785,
      "language_loss": 0.73796386,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76055306,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 2.7751646041870117
    },
    {
      "auxiliary_loss_clip": 0.01221942,
      "auxiliary_loss_mlp": 0.01040921,
      "balance_loss_clip": 0.95110178,
      "balance_loss_mlp": 1.02826154,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.0832857281675423,
      "language_loss": 0.86755133,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.89017993,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 2.746945381164551
    },
    {
      "auxiliary_loss_clip": 0.01240315,
      "auxiliary_loss_mlp": 0.01043261,
      "balance_loss_clip": 1.06866193,
      "balance_loss_mlp": 1.03144109,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 1.8939059257850888,
      "language_loss": 0.70128417,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.7241199,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.765996217727661
    },
    {
      "auxiliary_loss_clip": 0.0121467,
      "auxiliary_loss_mlp": 0.01035,
      "balance_loss_clip": 0.94882143,
      "balance_loss_mlp": 1.02286434,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 2.311881755721559,
      "language_loss": 0.84485173,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.86734843,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 2.827521800994873
    },
    {
      "auxiliary_loss_clip": 0.01226881,
      "auxiliary_loss_mlp": 0.01047538,
      "balance_loss_clip": 0.98539865,
      "balance_loss_mlp": 1.0347352,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 1.6811740354003715,
      "language_loss": 0.84324551,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86598969,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 2.7346200942993164
    },
    {
      "auxiliary_loss_clip": 0.01231754,
      "auxiliary_loss_mlp": 0.0104432,
      "balance_loss_clip": 0.95199001,
      "balance_loss_mlp": 1.03259003,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 2.933112257532739,
      "language_loss": 0.76062942,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78339016,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 4.210619688034058
    },
    {
      "auxiliary_loss_clip": 0.01224522,
      "auxiliary_loss_mlp": 0.01036336,
      "balance_loss_clip": 0.98995185,
      "balance_loss_mlp": 1.02480793,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.5945946944379061,
      "language_loss": 0.75177598,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.7743845,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 4.6529927253723145
    },
    {
      "auxiliary_loss_clip": 0.01227095,
      "auxiliary_loss_mlp": 0.01040667,
      "balance_loss_clip": 1.02759409,
      "balance_loss_mlp": 1.02891934,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 1.8010721793572988,
      "language_loss": 0.90759695,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.93027461,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 3.6907455921173096
    },
    {
      "auxiliary_loss_clip": 0.01225909,
      "auxiliary_loss_mlp": 0.01035829,
      "balance_loss_clip": 0.91374052,
      "balance_loss_mlp": 1.02429008,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.5829918972225467,
      "language_loss": 0.79250598,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.81512332,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.7145824432373047
    },
    {
      "auxiliary_loss_clip": 0.0121635,
      "auxiliary_loss_mlp": 0.01041087,
      "balance_loss_clip": 0.9072448,
      "balance_loss_mlp": 1.02961874,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.082071215423518,
      "language_loss": 0.84159958,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.86417395,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 2.837521553039551
    },
    {
      "auxiliary_loss_clip": 0.01238828,
      "auxiliary_loss_mlp": 0.01035183,
      "balance_loss_clip": 1.07068753,
      "balance_loss_mlp": 1.02346504,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 2.0003361935031454,
      "language_loss": 0.82137144,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84411156,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.5987327098846436
    },
    {
      "auxiliary_loss_clip": 0.01235052,
      "auxiliary_loss_mlp": 0.01043091,
      "balance_loss_clip": 1.0667038,
      "balance_loss_mlp": 1.03066325,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 1.8627974949670316,
      "language_loss": 0.83480388,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85758531,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.668107032775879
    },
    {
      "auxiliary_loss_clip": 0.01232208,
      "auxiliary_loss_mlp": 0.01043434,
      "balance_loss_clip": 0.99076164,
      "balance_loss_mlp": 1.03131032,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 3.168099619445142,
      "language_loss": 0.87245548,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.89521182,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.782545566558838
    },
    {
      "auxiliary_loss_clip": 0.01233386,
      "auxiliary_loss_mlp": 0.0104458,
      "balance_loss_clip": 1.02753139,
      "balance_loss_mlp": 1.03233743,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 1.8983468749217998,
      "language_loss": 0.69871825,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72149789,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 2.8548550605773926
    },
    {
      "auxiliary_loss_clip": 0.01232128,
      "auxiliary_loss_mlp": 0.01128901,
      "balance_loss_clip": 1.02651024,
      "balance_loss_mlp": 0.0,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 1.7529941432230547,
      "language_loss": 0.86095941,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.88456959,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.7659494876861572
    },
    {
      "auxiliary_loss_clip": 0.01232796,
      "auxiliary_loss_mlp": 0.01048101,
      "balance_loss_clip": 0.98719925,
      "balance_loss_mlp": 1.03579843,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 1.9471018818295216,
      "language_loss": 0.77537262,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79818165,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.7901663780212402
    },
    {
      "auxiliary_loss_clip": 0.01234342,
      "auxiliary_loss_mlp": 0.01045783,
      "balance_loss_clip": 1.02642131,
      "balance_loss_mlp": 1.03305149,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 2.8411161584535125,
      "language_loss": 0.70637488,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72917616,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.7428128719329834
    },
    {
      "auxiliary_loss_clip": 0.01199616,
      "auxiliary_loss_mlp": 0.01045881,
      "balance_loss_clip": 0.94199896,
      "balance_loss_mlp": 1.03351867,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 1.9083550493110222,
      "language_loss": 0.79177034,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81422526,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.681762933731079
    },
    {
      "auxiliary_loss_clip": 0.01227082,
      "auxiliary_loss_mlp": 0.01043801,
      "balance_loss_clip": 0.98481101,
      "balance_loss_mlp": 1.03216636,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 1.9293839154283492,
      "language_loss": 0.71752781,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.74023664,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.7709639072418213
    },
    {
      "auxiliary_loss_clip": 0.01220997,
      "auxiliary_loss_mlp": 0.01039126,
      "balance_loss_clip": 0.90959537,
      "balance_loss_mlp": 1.02783084,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.7674736667931774,
      "language_loss": 0.81226337,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.83486462,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.9140021800994873
    },
    {
      "auxiliary_loss_clip": 0.01237489,
      "auxiliary_loss_mlp": 0.01051125,
      "balance_loss_clip": 1.02844596,
      "balance_loss_mlp": 1.03811884,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 1.6780898199763552,
      "language_loss": 0.83128405,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.8541702,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.918734073638916
    },
    {
      "auxiliary_loss_clip": 0.0112882,
      "auxiliary_loss_mlp": 0.01006281,
      "balance_loss_clip": 0.92935562,
      "balance_loss_mlp": 0.99965346,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.8530619383006204,
      "language_loss": 0.64774299,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66909397,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.4279837608337402
    },
    {
      "auxiliary_loss_clip": 0.01227739,
      "auxiliary_loss_mlp": 0.01044888,
      "balance_loss_clip": 0.94888252,
      "balance_loss_mlp": 1.03398669,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.7833204429331702,
      "language_loss": 0.79831445,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.82104069,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.736335039138794
    },
    {
      "auxiliary_loss_clip": 0.01224793,
      "auxiliary_loss_mlp": 0.01041883,
      "balance_loss_clip": 0.99055153,
      "balance_loss_mlp": 1.02922916,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 1.6632860183262241,
      "language_loss": 0.85205984,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87472653,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.9242546558380127
    },
    {
      "auxiliary_loss_clip": 0.01231811,
      "auxiliary_loss_mlp": 0.01045354,
      "balance_loss_clip": 1.02461755,
      "balance_loss_mlp": 1.03212762,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 1.7682526914397703,
      "language_loss": 0.85192329,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87469494,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.6849324703216553
    },
    {
      "auxiliary_loss_clip": 0.01227406,
      "auxiliary_loss_mlp": 0.0104223,
      "balance_loss_clip": 0.91015702,
      "balance_loss_mlp": 1.03021955,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 2.0789535501356196,
      "language_loss": 0.83676517,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.85946155,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 3.0056419372558594
    },
    {
      "auxiliary_loss_clip": 0.01234204,
      "auxiliary_loss_mlp": 0.010387,
      "balance_loss_clip": 1.06396103,
      "balance_loss_mlp": 1.02677917,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.9850027306982112,
      "language_loss": 0.78704464,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.80977368,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.711754560470581
    },
    {
      "auxiliary_loss_clip": 0.01222026,
      "auxiliary_loss_mlp": 0.0112972,
      "balance_loss_clip": 1.02154672,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 2.083513749451306,
      "language_loss": 0.88063622,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.9041537,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 2.6906468868255615
    },
    {
      "auxiliary_loss_clip": 0.01239353,
      "auxiliary_loss_mlp": 0.01129338,
      "balance_loss_clip": 1.02754235,
      "balance_loss_mlp": 0.0,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.601970722078597,
      "language_loss": 0.80351412,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82720101,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.8243651390075684
    },
    {
      "auxiliary_loss_clip": 0.0123627,
      "auxiliary_loss_mlp": 0.01046701,
      "balance_loss_clip": 0.94588006,
      "balance_loss_mlp": 1.03390408,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 2.1155828470949363,
      "language_loss": 0.83105707,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.85388678,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 2.7223563194274902
    },
    {
      "auxiliary_loss_clip": 0.01230302,
      "auxiliary_loss_mlp": 0.0104282,
      "balance_loss_clip": 0.99252617,
      "balance_loss_mlp": 1.03128672,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 3.1799031612485518,
      "language_loss": 0.82584751,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.84857869,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.7354133129119873
    },
    {
      "auxiliary_loss_clip": 0.01236457,
      "auxiliary_loss_mlp": 0.01044647,
      "balance_loss_clip": 1.02719879,
      "balance_loss_mlp": 1.03191507,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 2.6709572220624787,
      "language_loss": 0.68820655,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.71101755,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 4.857868909835815
    },
    {
      "auxiliary_loss_clip": 0.01212285,
      "auxiliary_loss_mlp": 0.01048389,
      "balance_loss_clip": 0.94265711,
      "balance_loss_mlp": 1.03584194,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 1.880410760292643,
      "language_loss": 0.86905593,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.89166272,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 3.6649887561798096
    },
    {
      "auxiliary_loss_clip": 0.01224096,
      "auxiliary_loss_mlp": 0.01040315,
      "balance_loss_clip": 0.94626725,
      "balance_loss_mlp": 1.02773869,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 1.7865717118075302,
      "language_loss": 0.75735497,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.77999908,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 3.762666940689087
    },
    {
      "auxiliary_loss_clip": 0.01239143,
      "auxiliary_loss_mlp": 0.01040221,
      "balance_loss_clip": 1.0291748,
      "balance_loss_mlp": 1.02720308,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 2.5112205189004726,
      "language_loss": 0.93775666,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.96055037,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 2.748765468597412
    },
    {
      "auxiliary_loss_clip": 0.0123292,
      "auxiliary_loss_mlp": 0.01052046,
      "balance_loss_clip": 0.95149064,
      "balance_loss_mlp": 1.03896832,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 2.3024234663252106,
      "language_loss": 0.81976414,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.84261382,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 2.841552972793579
    },
    {
      "auxiliary_loss_clip": 0.0123209,
      "auxiliary_loss_mlp": 0.01047999,
      "balance_loss_clip": 0.95314968,
      "balance_loss_mlp": 1.03511822,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 2.148075943429343,
      "language_loss": 0.76296067,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.78576159,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.7504730224609375
    },
    {
      "auxiliary_loss_clip": 0.01233114,
      "auxiliary_loss_mlp": 0.010522,
      "balance_loss_clip": 1.02763677,
      "balance_loss_mlp": 1.03869367,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 1.7109608932952798,
      "language_loss": 0.78138393,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.80423701,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.710857629776001
    },
    {
      "auxiliary_loss_clip": 0.01121227,
      "auxiliary_loss_mlp": 0.01011628,
      "balance_loss_clip": 0.92059386,
      "balance_loss_mlp": 1.00538123,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7671432061988768,
      "language_loss": 0.55671811,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.57804668,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.2900147438049316
    },
    {
      "auxiliary_loss_clip": 0.01237119,
      "auxiliary_loss_mlp": 0.01038355,
      "balance_loss_clip": 1.02412748,
      "balance_loss_mlp": 1.02549815,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 2.171342587209219,
      "language_loss": 0.7133835,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73613822,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.8074417114257812
    },
    {
      "auxiliary_loss_clip": 0.01221075,
      "auxiliary_loss_mlp": 0.01045533,
      "balance_loss_clip": 1.01961005,
      "balance_loss_mlp": 1.03301585,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.9153391343439314,
      "language_loss": 0.7427392,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.7654053,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.7628114223480225
    },
    {
      "auxiliary_loss_clip": 0.01228902,
      "auxiliary_loss_mlp": 0.01047142,
      "balance_loss_clip": 1.02458525,
      "balance_loss_mlp": 1.03460121,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 2.309797097935536,
      "language_loss": 0.72327018,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.74603063,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.7211568355560303
    },
    {
      "auxiliary_loss_clip": 0.0122806,
      "auxiliary_loss_mlp": 0.01039337,
      "balance_loss_clip": 1.02630663,
      "balance_loss_mlp": 1.02662385,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 1.758258228518568,
      "language_loss": 0.9668386,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.98951256,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.7194104194641113
    },
    {
      "auxiliary_loss_clip": 0.01218394,
      "auxiliary_loss_mlp": 0.01043698,
      "balance_loss_clip": 0.98250741,
      "balance_loss_mlp": 1.0317775,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 3.7799219353928115,
      "language_loss": 0.85625136,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.87887233,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.716390371322632
    },
    {
      "auxiliary_loss_clip": 0.01234278,
      "auxiliary_loss_mlp": 0.01047257,
      "balance_loss_clip": 1.06496596,
      "balance_loss_mlp": 1.03496051,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 1.9085672999010814,
      "language_loss": 0.8135708,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.83638608,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.694099187850952
    },
    {
      "auxiliary_loss_clip": 0.01219761,
      "auxiliary_loss_mlp": 0.01041862,
      "balance_loss_clip": 0.94771725,
      "balance_loss_mlp": 1.02950573,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 1.763546391387733,
      "language_loss": 0.79916131,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.82177758,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.8142940998077393
    },
    {
      "auxiliary_loss_clip": 0.0123615,
      "auxiliary_loss_mlp": 0.01042132,
      "balance_loss_clip": 0.98531479,
      "balance_loss_mlp": 1.0288105,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 2.6172135906587606,
      "language_loss": 0.64679909,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.66958195,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.7478079795837402
    },
    {
      "auxiliary_loss_clip": 0.01230833,
      "auxiliary_loss_mlp": 0.01044476,
      "balance_loss_clip": 1.06592572,
      "balance_loss_mlp": 1.03303838,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 1.6466897164928096,
      "language_loss": 0.85094035,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87369347,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.7426059246063232
    },
    {
      "auxiliary_loss_clip": 0.01220643,
      "auxiliary_loss_mlp": 0.01043509,
      "balance_loss_clip": 0.83200169,
      "balance_loss_mlp": 1.03131986,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 2.2658592817786687,
      "language_loss": 0.81105995,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83370149,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 3.0602848529815674
    },
    {
      "auxiliary_loss_clip": 0.0121993,
      "auxiliary_loss_mlp": 0.0104488,
      "balance_loss_clip": 1.0217396,
      "balance_loss_mlp": 1.03189206,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 1.8953585622275928,
      "language_loss": 0.8206417,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84328979,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.9852817058563232
    },
    {
      "auxiliary_loss_clip": 0.01205645,
      "auxiliary_loss_mlp": 0.01041268,
      "balance_loss_clip": 0.90482986,
      "balance_loss_mlp": 1.02840519,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 1.8824117680012327,
      "language_loss": 0.65039021,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.67285937,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.8784945011138916
    },
    {
      "auxiliary_loss_clip": 0.01233315,
      "auxiliary_loss_mlp": 0.01041885,
      "balance_loss_clip": 0.94633901,
      "balance_loss_mlp": 1.02938557,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 2.3233128328616037,
      "language_loss": 0.89540172,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91815376,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.854956865310669
    },
    {
      "auxiliary_loss_clip": 0.01225897,
      "auxiliary_loss_mlp": 0.01037218,
      "balance_loss_clip": 1.02572107,
      "balance_loss_mlp": 1.02440274,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.7934926377506148,
      "language_loss": 0.8219499,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84458101,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 2.869370937347412
    },
    {
      "auxiliary_loss_clip": 0.01225722,
      "auxiliary_loss_mlp": 0.01045402,
      "balance_loss_clip": 0.98793972,
      "balance_loss_mlp": 1.03261709,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.8931316025083595,
      "language_loss": 0.80484235,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82755363,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.774563789367676
    },
    {
      "auxiliary_loss_clip": 0.01213628,
      "auxiliary_loss_mlp": 0.01036461,
      "balance_loss_clip": 0.82751083,
      "balance_loss_mlp": 1.0244031,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 1.9406424695908273,
      "language_loss": 0.69160384,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71410477,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 2.9536516666412354
    },
    {
      "auxiliary_loss_clip": 0.01231829,
      "auxiliary_loss_mlp": 0.0103817,
      "balance_loss_clip": 0.98635006,
      "balance_loss_mlp": 1.0258559,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.6678856268146347,
      "language_loss": 0.77938604,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80208606,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 3.2001824378967285
    },
    {
      "auxiliary_loss_clip": 0.01229589,
      "auxiliary_loss_mlp": 0.01128967,
      "balance_loss_clip": 0.99180132,
      "balance_loss_mlp": 0.0,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 8.837652653515654,
      "language_loss": 0.78506064,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.8086462,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 2.853957414627075
    },
    {
      "auxiliary_loss_clip": 0.01212236,
      "auxiliary_loss_mlp": 0.01035318,
      "balance_loss_clip": 0.94677943,
      "balance_loss_mlp": 1.02297401,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.289679515863155,
      "language_loss": 0.78395879,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80643427,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 3.942830801010132
    },
    {
      "auxiliary_loss_clip": 0.01233982,
      "auxiliary_loss_mlp": 0.01129234,
      "balance_loss_clip": 1.06310439,
      "balance_loss_mlp": 0.0,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.8339305502273286,
      "language_loss": 0.71477395,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.73840612,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 3.6408591270446777
    },
    {
      "auxiliary_loss_clip": 0.0123567,
      "auxiliary_loss_mlp": 0.01043266,
      "balance_loss_clip": 1.06557274,
      "balance_loss_mlp": 1.0312264,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 1.983929953568458,
      "language_loss": 0.7529037,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.77569306,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 3.652254581451416
    },
    {
      "auxiliary_loss_clip": 0.01216464,
      "auxiliary_loss_mlp": 0.0105073,
      "balance_loss_clip": 0.97976363,
      "balance_loss_mlp": 1.03870177,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 2.4230533222140234,
      "language_loss": 0.89756083,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.92023277,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 2.8154423236846924
    },
    {
      "auxiliary_loss_clip": 0.01224658,
      "auxiliary_loss_mlp": 0.01041273,
      "balance_loss_clip": 0.98580194,
      "balance_loss_mlp": 1.02881551,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 3.0224923476707533,
      "language_loss": 0.85999954,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.88265884,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.8087003231048584
    },
    {
      "auxiliary_loss_clip": 0.01236821,
      "auxiliary_loss_mlp": 0.01044043,
      "balance_loss_clip": 1.06781828,
      "balance_loss_mlp": 1.0321579,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.1649216905026045,
      "language_loss": 0.73580289,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.7586115,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.7245981693267822
    },
    {
      "auxiliary_loss_clip": 0.01235149,
      "auxiliary_loss_mlp": 0.0112891,
      "balance_loss_clip": 1.06429625,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 5.757984740221322,
      "language_loss": 0.85460615,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.87824678,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.7068254947662354
    },
    {
      "auxiliary_loss_clip": 0.01229552,
      "auxiliary_loss_mlp": 0.01040866,
      "balance_loss_clip": 1.02970576,
      "balance_loss_mlp": 1.02923083,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 3.061823811143228,
      "language_loss": 0.77205956,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.79476374,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.674532651901245
    },
    {
      "auxiliary_loss_clip": 0.0122931,
      "auxiliary_loss_mlp": 0.01046056,
      "balance_loss_clip": 1.02655315,
      "balance_loss_mlp": 1.0340755,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 2.2038799514953564,
      "language_loss": 0.84365523,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86640888,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.764723062515259
    },
    {
      "auxiliary_loss_clip": 0.01228417,
      "auxiliary_loss_mlp": 0.01042838,
      "balance_loss_clip": 1.02750301,
      "balance_loss_mlp": 1.03120351,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 2.1122990409795133,
      "language_loss": 0.7508167,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77352929,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.742495536804199
    },
    {
      "auxiliary_loss_clip": 0.0122201,
      "auxiliary_loss_mlp": 0.01033416,
      "balance_loss_clip": 0.9483723,
      "balance_loss_mlp": 1.02236557,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 1.7149227423759787,
      "language_loss": 0.88839769,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.91095197,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.8267555236816406
    },
    {
      "auxiliary_loss_clip": 0.01235088,
      "auxiliary_loss_mlp": 0.01051498,
      "balance_loss_clip": 1.02888596,
      "balance_loss_mlp": 1.03920734,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 3.9496833556694173,
      "language_loss": 0.83667713,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.85954297,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.737396240234375
    },
    {
      "auxiliary_loss_clip": 0.01228297,
      "auxiliary_loss_mlp": 0.01038308,
      "balance_loss_clip": 1.02552533,
      "balance_loss_mlp": 1.02666688,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.728345288882742,
      "language_loss": 0.74786675,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.77053273,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.887930393218994
    },
    {
      "auxiliary_loss_clip": 0.0123362,
      "auxiliary_loss_mlp": 0.01039017,
      "balance_loss_clip": 1.06557298,
      "balance_loss_mlp": 1.02770424,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 1.517728823874865,
      "language_loss": 0.78431201,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.80703843,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.6822500228881836
    },
    {
      "auxiliary_loss_clip": 0.01233668,
      "auxiliary_loss_mlp": 0.01048255,
      "balance_loss_clip": 1.06409097,
      "balance_loss_mlp": 1.03587556,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 2.168974992994287,
      "language_loss": 0.70439959,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.72721887,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.5870983600616455
    },
    {
      "auxiliary_loss_clip": 0.01228572,
      "auxiliary_loss_mlp": 0.01043198,
      "balance_loss_clip": 0.94999838,
      "balance_loss_mlp": 1.03042507,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.9454129117768217,
      "language_loss": 0.97092819,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.99364597,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.682025194168091
    },
    {
      "auxiliary_loss_clip": 0.01220847,
      "auxiliary_loss_mlp": 0.01042242,
      "balance_loss_clip": 0.98541236,
      "balance_loss_mlp": 1.03062487,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 2.242179404391003,
      "language_loss": 0.73716825,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.75979912,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.642958879470825
    },
    {
      "auxiliary_loss_clip": 0.01234468,
      "auxiliary_loss_mlp": 0.01034511,
      "balance_loss_clip": 1.02860308,
      "balance_loss_mlp": 1.0231626,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 3.0559080999358943,
      "language_loss": 0.7801221,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.80281186,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.5828306674957275
    },
    {
      "auxiliary_loss_clip": 0.01230786,
      "auxiliary_loss_mlp": 0.01034195,
      "balance_loss_clip": 0.98569381,
      "balance_loss_mlp": 1.02190471,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 1.8770510841364272,
      "language_loss": 0.7386142,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.76126397,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.6467573642730713
    },
    {
      "auxiliary_loss_clip": 0.01235069,
      "auxiliary_loss_mlp": 0.01047225,
      "balance_loss_clip": 1.06623495,
      "balance_loss_mlp": 1.03564441,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 1.7853944121593608,
      "language_loss": 0.84207112,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86489403,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.613157033920288
    },
    {
      "auxiliary_loss_clip": 0.012285,
      "auxiliary_loss_mlp": 0.01038176,
      "balance_loss_clip": 0.94664395,
      "balance_loss_mlp": 1.02630305,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 2.0886725112859303,
      "language_loss": 0.64578456,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.66845131,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 2.723633289337158
    },
    {
      "auxiliary_loss_clip": 0.01224018,
      "auxiliary_loss_mlp": 0.01035788,
      "balance_loss_clip": 0.9891625,
      "balance_loss_mlp": 1.02488089,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 2.2658890327167924,
      "language_loss": 0.79625148,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.81884956,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 2.6491146087646484
    },
    {
      "auxiliary_loss_clip": 0.01233557,
      "auxiliary_loss_mlp": 0.01128699,
      "balance_loss_clip": 1.02795815,
      "balance_loss_mlp": 0.0,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.1008383311213406,
      "language_loss": 0.82173777,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.84536028,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 2.627596616744995
    },
    {
      "auxiliary_loss_clip": 0.01233174,
      "auxiliary_loss_mlp": 0.01036935,
      "balance_loss_clip": 1.06664348,
      "balance_loss_mlp": 1.02484739,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 2.6130040472406897,
      "language_loss": 0.7033968,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72609788,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.755383253097534
    },
    {
      "auxiliary_loss_clip": 0.01223051,
      "auxiliary_loss_mlp": 0.01038917,
      "balance_loss_clip": 1.02611125,
      "balance_loss_mlp": 1.02755046,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 2.4196431347032585,
      "language_loss": 0.89976943,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.92238903,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.617854118347168
    },
    {
      "auxiliary_loss_clip": 0.01233858,
      "auxiliary_loss_mlp": 0.01040756,
      "balance_loss_clip": 1.02981055,
      "balance_loss_mlp": 1.02838755,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 2.11431975963493,
      "language_loss": 0.83568048,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.85842657,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.6128957271575928
    },
    {
      "auxiliary_loss_clip": 0.01224135,
      "auxiliary_loss_mlp": 0.01034492,
      "balance_loss_clip": 0.9846819,
      "balance_loss_mlp": 1.02297056,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 2.070111295788625,
      "language_loss": 0.72765291,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.75023913,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 4.590798377990723
    },
    {
      "auxiliary_loss_clip": 0.01216886,
      "auxiliary_loss_mlp": 0.0103708,
      "balance_loss_clip": 0.90453827,
      "balance_loss_mlp": 1.02540922,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 2.8432337311121443,
      "language_loss": 0.70153594,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.72407562,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 3.5294976234436035
    },
    {
      "auxiliary_loss_clip": 0.01213317,
      "auxiliary_loss_mlp": 0.01048006,
      "balance_loss_clip": 0.86826265,
      "balance_loss_mlp": 1.03634143,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 3.5671324649027465,
      "language_loss": 0.89938664,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92199993,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 3.7408287525177
    },
    {
      "auxiliary_loss_clip": 0.01232232,
      "auxiliary_loss_mlp": 0.0104268,
      "balance_loss_clip": 0.98898506,
      "balance_loss_mlp": 1.03057468,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 2.4194191791030537,
      "language_loss": 0.82190394,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.84465313,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.6756644248962402
    },
    {
      "auxiliary_loss_clip": 0.01225706,
      "auxiliary_loss_mlp": 0.01041218,
      "balance_loss_clip": 0.9422909,
      "balance_loss_mlp": 1.02923751,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 7.381072012183392,
      "language_loss": 0.78219306,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80486226,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.880070924758911
    },
    {
      "auxiliary_loss_clip": 0.01232322,
      "auxiliary_loss_mlp": 0.01043362,
      "balance_loss_clip": 1.02671731,
      "balance_loss_mlp": 1.03203726,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.4949828831273027,
      "language_loss": 0.83492076,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85767758,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 2.77994704246521
    },
    {
      "auxiliary_loss_clip": 0.01229412,
      "auxiliary_loss_mlp": 0.01036665,
      "balance_loss_clip": 0.98805285,
      "balance_loss_mlp": 1.02595401,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.0258291403406266,
      "language_loss": 0.76237994,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78504068,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.6642942428588867
    },
    {
      "auxiliary_loss_clip": 0.01228519,
      "auxiliary_loss_mlp": 0.01033426,
      "balance_loss_clip": 1.02469063,
      "balance_loss_mlp": 1.02161837,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.1604243589624463,
      "language_loss": 0.82430255,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84692204,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.6723389625549316
    },
    {
      "auxiliary_loss_clip": 0.01226276,
      "auxiliary_loss_mlp": 0.01035107,
      "balance_loss_clip": 1.02504897,
      "balance_loss_mlp": 1.02375209,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 2.248452549651491,
      "language_loss": 0.82230866,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.84492254,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.7076826095581055
    },
    {
      "auxiliary_loss_clip": 0.01212332,
      "auxiliary_loss_mlp": 0.01035615,
      "balance_loss_clip": 0.94428015,
      "balance_loss_mlp": 1.02485013,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 2.1471548554948874,
      "language_loss": 0.88015819,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.9026376,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.682464361190796
    },
    {
      "auxiliary_loss_clip": 0.01225325,
      "auxiliary_loss_mlp": 0.01033729,
      "balance_loss_clip": 0.90851253,
      "balance_loss_mlp": 1.02205825,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 2.4096376149682404,
      "language_loss": 0.87028837,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.89287889,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 2.752952814102173
    },
    {
      "auxiliary_loss_clip": 0.01226591,
      "auxiliary_loss_mlp": 0.01049818,
      "balance_loss_clip": 0.98437667,
      "balance_loss_mlp": 1.03875518,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.76426439244477,
      "language_loss": 0.74600625,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.76877034,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.68622088432312
    },
    {
      "auxiliary_loss_clip": 0.01234935,
      "auxiliary_loss_mlp": 0.01050071,
      "balance_loss_clip": 1.06854463,
      "balance_loss_mlp": 1.03806674,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.849062817960655,
      "language_loss": 0.82637578,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.84922588,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.6700174808502197
    },
    {
      "auxiliary_loss_clip": 0.0121307,
      "auxiliary_loss_mlp": 0.01041421,
      "balance_loss_clip": 0.94237214,
      "balance_loss_mlp": 1.02850497,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.2766283184056926,
      "language_loss": 0.90532613,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.92787105,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.733867883682251
    },
    {
      "auxiliary_loss_clip": 0.01231905,
      "auxiliary_loss_mlp": 0.01128948,
      "balance_loss_clip": 1.02485704,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 1.9313372982048622,
      "language_loss": 0.7772783,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.80088687,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.736259698867798
    },
    {
      "auxiliary_loss_clip": 0.01224474,
      "auxiliary_loss_mlp": 0.01040938,
      "balance_loss_clip": 0.98979008,
      "balance_loss_mlp": 1.0296489,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 1.846768637217415,
      "language_loss": 0.8579855,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88063961,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.6625282764434814
    },
    {
      "auxiliary_loss_clip": 0.01224553,
      "auxiliary_loss_mlp": 0.01037764,
      "balance_loss_clip": 0.98844635,
      "balance_loss_mlp": 1.02586102,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 2.0508242991571706,
      "language_loss": 0.79421937,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81684256,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.7617204189300537
    },
    {
      "auxiliary_loss_clip": 0.0122681,
      "auxiliary_loss_mlp": 0.01035389,
      "balance_loss_clip": 0.95150286,
      "balance_loss_mlp": 1.02323532,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 2.394953443484635,
      "language_loss": 0.72260803,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.74523002,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 2.9862589836120605
    },
    {
      "auxiliary_loss_clip": 0.01232056,
      "auxiliary_loss_mlp": 0.01037627,
      "balance_loss_clip": 1.02637339,
      "balance_loss_mlp": 1.02623713,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 1.9741995416251255,
      "language_loss": 0.94095552,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.96365237,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.66083025932312
    },
    {
      "auxiliary_loss_clip": 0.01210205,
      "auxiliary_loss_mlp": 0.01042659,
      "balance_loss_clip": 0.8269667,
      "balance_loss_mlp": 1.03084588,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 1.773315054978524,
      "language_loss": 0.88485491,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90738356,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.9048283100128174
    },
    {
      "auxiliary_loss_clip": 0.01210796,
      "auxiliary_loss_mlp": 0.01039214,
      "balance_loss_clip": 0.98251009,
      "balance_loss_mlp": 1.02793646,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 2.0118344211743797,
      "language_loss": 0.89131862,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.91381872,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.8416314125061035
    },
    {
      "auxiliary_loss_clip": 0.01205727,
      "auxiliary_loss_mlp": 0.01034508,
      "balance_loss_clip": 0.98139668,
      "balance_loss_mlp": 1.02208614,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 1.7816771621312106,
      "language_loss": 0.88499963,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90740198,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 2.732807159423828
    },
    {
      "auxiliary_loss_clip": 0.01229017,
      "auxiliary_loss_mlp": 0.01043282,
      "balance_loss_clip": 0.94801986,
      "balance_loss_mlp": 1.03203452,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 1.967113094709023,
      "language_loss": 0.73256457,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75528753,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.77341890335083
    },
    {
      "auxiliary_loss_clip": 0.01234834,
      "auxiliary_loss_mlp": 0.01045069,
      "balance_loss_clip": 1.02972674,
      "balance_loss_mlp": 1.03301752,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 1.8611343135871306,
      "language_loss": 0.77682149,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.79962051,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.701951742172241
    },
    {
      "auxiliary_loss_clip": 0.01221785,
      "auxiliary_loss_mlp": 0.01036257,
      "balance_loss_clip": 1.02302837,
      "balance_loss_mlp": 1.02509332,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.9149205295622154,
      "language_loss": 0.82557416,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.84815466,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 2.8422696590423584
    },
    {
      "auxiliary_loss_clip": 0.01225388,
      "auxiliary_loss_mlp": 0.01037768,
      "balance_loss_clip": 0.98726863,
      "balance_loss_mlp": 1.02647924,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 1.8934625899656476,
      "language_loss": 0.84825993,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.87089157,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.7545578479766846
    },
    {
      "auxiliary_loss_clip": 0.01224183,
      "auxiliary_loss_mlp": 0.01039389,
      "balance_loss_clip": 0.90951014,
      "balance_loss_mlp": 1.02793932,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 2.179173887091971,
      "language_loss": 0.76404941,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78668517,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 5.005924224853516
    },
    {
      "auxiliary_loss_clip": 0.01227395,
      "auxiliary_loss_mlp": 0.01036862,
      "balance_loss_clip": 1.02330709,
      "balance_loss_mlp": 1.02526283,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 10.380757658990534,
      "language_loss": 0.74258733,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.76522988,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 3.631434679031372
    },
    {
      "auxiliary_loss_clip": 0.01228097,
      "auxiliary_loss_mlp": 0.01036697,
      "balance_loss_clip": 0.988478,
      "balance_loss_mlp": 1.02497315,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 1.9457665714382548,
      "language_loss": 0.77623892,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.7988869,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 3.5984115600585938
    },
    {
      "auxiliary_loss_clip": 0.01226995,
      "auxiliary_loss_mlp": 0.01037333,
      "balance_loss_clip": 1.02721763,
      "balance_loss_mlp": 1.02608609,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 1.6541652061195207,
      "language_loss": 0.75930393,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.7819472,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.748771905899048
    },
    {
      "auxiliary_loss_clip": 0.01223937,
      "auxiliary_loss_mlp": 0.01040728,
      "balance_loss_clip": 0.98688316,
      "balance_loss_mlp": 1.02921844,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 1.9188733986793403,
      "language_loss": 0.86372954,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.88637626,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.6719393730163574
    },
    {
      "auxiliary_loss_clip": 0.01235108,
      "auxiliary_loss_mlp": 0.01043203,
      "balance_loss_clip": 1.0324651,
      "balance_loss_mlp": 1.03122854,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 1.9349032104798722,
      "language_loss": 0.80322105,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.82600421,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 2.723912000656128
    },
    {
      "auxiliary_loss_clip": 0.01218309,
      "auxiliary_loss_mlp": 0.01047734,
      "balance_loss_clip": 0.98466122,
      "balance_loss_mlp": 1.03594458,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 1.4438862020939858,
      "language_loss": 0.76688373,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.78954411,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.795224905014038
    },
    {
      "auxiliary_loss_clip": 0.01217968,
      "auxiliary_loss_mlp": 0.01037956,
      "balance_loss_clip": 1.02297032,
      "balance_loss_mlp": 1.02631497,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 2.257720894285293,
      "language_loss": 0.8604728,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88303208,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.5969855785369873
    },
    {
      "auxiliary_loss_clip": 0.01220485,
      "auxiliary_loss_mlp": 0.01037635,
      "balance_loss_clip": 0.98606718,
      "balance_loss_mlp": 1.02599478,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 1.8813768281420014,
      "language_loss": 0.82383269,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.84641391,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.737710475921631
    },
    {
      "auxiliary_loss_clip": 0.0123031,
      "auxiliary_loss_mlp": 0.01038114,
      "balance_loss_clip": 0.90758467,
      "balance_loss_mlp": 1.02649689,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 1.762893373672034,
      "language_loss": 0.79635376,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81903803,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.760601758956909
    },
    {
      "auxiliary_loss_clip": 0.01143787,
      "auxiliary_loss_mlp": 0.01012285,
      "balance_loss_clip": 1.00979996,
      "balance_loss_mlp": 1.00599122,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8868631259676234,
      "language_loss": 0.59230995,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61387068,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.2639870643615723
    },
    {
      "auxiliary_loss_clip": 0.01228282,
      "auxiliary_loss_mlp": 0.01043531,
      "balance_loss_clip": 1.02609825,
      "balance_loss_mlp": 1.03212881,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.7626170465810485,
      "language_loss": 0.74838138,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.77109957,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.7913713455200195
    },
    {
      "auxiliary_loss_clip": 0.01223749,
      "auxiliary_loss_mlp": 0.0103781,
      "balance_loss_clip": 0.99046838,
      "balance_loss_mlp": 1.02629471,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 2.560326297936765,
      "language_loss": 0.84168375,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.8642993,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.7548065185546875
    },
    {
      "auxiliary_loss_clip": 0.01232699,
      "auxiliary_loss_mlp": 0.01041541,
      "balance_loss_clip": 1.02446342,
      "balance_loss_mlp": 1.02994752,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 3.541023487698327,
      "language_loss": 0.62166244,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.64440489,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.722886800765991
    },
    {
      "auxiliary_loss_clip": 0.01231502,
      "auxiliary_loss_mlp": 0.01037995,
      "balance_loss_clip": 0.99021524,
      "balance_loss_mlp": 1.02578795,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.4203257992358083,
      "language_loss": 0.86158156,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88427651,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.7167162895202637
    },
    {
      "auxiliary_loss_clip": 0.01219442,
      "auxiliary_loss_mlp": 0.01042471,
      "balance_loss_clip": 0.94876343,
      "balance_loss_mlp": 1.03151631,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 4.067680280230286,
      "language_loss": 0.70781565,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73043478,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.7123422622680664
    },
    {
      "auxiliary_loss_clip": 0.0123107,
      "auxiliary_loss_mlp": 0.01036743,
      "balance_loss_clip": 1.03033209,
      "balance_loss_mlp": 1.0251441,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.543598424540406,
      "language_loss": 0.87627077,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.89894891,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.6773979663848877
    },
    {
      "auxiliary_loss_clip": 0.01228331,
      "auxiliary_loss_mlp": 0.01034852,
      "balance_loss_clip": 1.06165028,
      "balance_loss_mlp": 1.02385449,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 9.32155719460005,
      "language_loss": 0.80993736,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83256912,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.80942964553833
    },
    {
      "auxiliary_loss_clip": 0.0122804,
      "auxiliary_loss_mlp": 0.0103624,
      "balance_loss_clip": 1.02629924,
      "balance_loss_mlp": 1.02499211,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 1.5051693959453383,
      "language_loss": 0.83437455,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.8570174,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.667391061782837
    },
    {
      "auxiliary_loss_clip": 0.01220826,
      "auxiliary_loss_mlp": 0.01035506,
      "balance_loss_clip": 0.98671401,
      "balance_loss_mlp": 1.02399611,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 2.861607890651191,
      "language_loss": 0.77954662,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80210996,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.69889497756958
    },
    {
      "auxiliary_loss_clip": 0.01233136,
      "auxiliary_loss_mlp": 0.01043412,
      "balance_loss_clip": 1.02864265,
      "balance_loss_mlp": 1.03118753,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 1.7265116724985607,
      "language_loss": 0.80947912,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.83224452,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.671769380569458
    },
    {
      "auxiliary_loss_clip": 0.01136759,
      "auxiliary_loss_mlp": 0.01004004,
      "balance_loss_clip": 0.92264867,
      "balance_loss_mlp": 0.99775785,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.8245697295824245,
      "language_loss": 0.61961877,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64102638,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 3.2462751865386963
    },
    {
      "auxiliary_loss_clip": 0.01230292,
      "auxiliary_loss_mlp": 0.01035399,
      "balance_loss_clip": 1.02896619,
      "balance_loss_mlp": 1.02437234,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 1.6645594369680958,
      "language_loss": 0.63924646,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66190338,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 2.746842622756958
    },
    {
      "auxiliary_loss_clip": 0.01230163,
      "auxiliary_loss_mlp": 0.0103986,
      "balance_loss_clip": 1.06424081,
      "balance_loss_mlp": 1.02869642,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 2.764081904499292,
      "language_loss": 0.80991924,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83261943,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.7216782569885254
    },
    {
      "auxiliary_loss_clip": 0.01226196,
      "auxiliary_loss_mlp": 0.01041579,
      "balance_loss_clip": 0.9067378,
      "balance_loss_mlp": 1.03000963,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 1.9635921029558314,
      "language_loss": 0.78114653,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80382431,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.777832269668579
    },
    {
      "auxiliary_loss_clip": 0.01212658,
      "auxiliary_loss_mlp": 0.01039339,
      "balance_loss_clip": 0.94743794,
      "balance_loss_mlp": 1.02792454,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 2.0623650898143815,
      "language_loss": 0.85619497,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.87871492,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.709420919418335
    },
    {
      "auxiliary_loss_clip": 0.01215993,
      "auxiliary_loss_mlp": 0.01045229,
      "balance_loss_clip": 0.90472209,
      "balance_loss_mlp": 1.03384435,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 2.7230034777646797,
      "language_loss": 0.82981384,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.85242605,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 4.940703392028809
    },
    {
      "auxiliary_loss_clip": 0.01234126,
      "auxiliary_loss_mlp": 0.01036902,
      "balance_loss_clip": 1.06716871,
      "balance_loss_mlp": 1.02479076,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 2.4278601317627437,
      "language_loss": 0.77180219,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.79451251,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 3.6280465126037598
    },
    {
      "auxiliary_loss_clip": 0.01211434,
      "auxiliary_loss_mlp": 0.01035232,
      "balance_loss_clip": 0.9806459,
      "balance_loss_mlp": 1.02364445,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 2.6480893325687536,
      "language_loss": 0.73900223,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.76146889,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 3.6154019832611084
    },
    {
      "auxiliary_loss_clip": 0.01232014,
      "auxiliary_loss_mlp": 0.01043236,
      "balance_loss_clip": 1.02474046,
      "balance_loss_mlp": 1.03186941,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 2.331320403433069,
      "language_loss": 0.77776033,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.80051279,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.7363672256469727
    },
    {
      "auxiliary_loss_clip": 0.01228452,
      "auxiliary_loss_mlp": 0.01051785,
      "balance_loss_clip": 1.02437329,
      "balance_loss_mlp": 1.03954184,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 2.046086523341349,
      "language_loss": 0.84964007,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87244236,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.7312135696411133
    },
    {
      "auxiliary_loss_clip": 0.012286,
      "auxiliary_loss_mlp": 0.01035124,
      "balance_loss_clip": 0.94967675,
      "balance_loss_mlp": 1.02366805,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 2.52196826343789,
      "language_loss": 0.89930749,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92194468,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.7508604526519775
    },
    {
      "auxiliary_loss_clip": 0.01230841,
      "auxiliary_loss_mlp": 0.01042762,
      "balance_loss_clip": 1.0661993,
      "balance_loss_mlp": 1.03119898,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 2.0784488097212788,
      "language_loss": 0.76569819,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.78843421,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 2.783740997314453
    },
    {
      "auxiliary_loss_clip": 0.01226953,
      "auxiliary_loss_mlp": 0.01048947,
      "balance_loss_clip": 1.02659631,
      "balance_loss_mlp": 1.03761601,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 1.931985383851977,
      "language_loss": 0.79058337,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.81334239,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.6750051975250244
    },
    {
      "auxiliary_loss_clip": 0.01224323,
      "auxiliary_loss_mlp": 0.01048175,
      "balance_loss_clip": 0.98691022,
      "balance_loss_mlp": 1.03608727,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 2.0034668027657347,
      "language_loss": 0.825086,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.84781098,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.7704145908355713
    },
    {
      "auxiliary_loss_clip": 0.0122769,
      "auxiliary_loss_mlp": 0.01040198,
      "balance_loss_clip": 1.02385712,
      "balance_loss_mlp": 1.02923727,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 2.3951314533643036,
      "language_loss": 0.77923679,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.80191565,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.6268343925476074
    },
    {
      "auxiliary_loss_clip": 0.01233229,
      "auxiliary_loss_mlp": 0.0103506,
      "balance_loss_clip": 1.02419484,
      "balance_loss_mlp": 1.0234257,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.0588487168097176,
      "language_loss": 0.84217793,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.86486089,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.6564981937408447
    },
    {
      "auxiliary_loss_clip": 0.01218089,
      "auxiliary_loss_mlp": 0.01043271,
      "balance_loss_clip": 0.94534445,
      "balance_loss_mlp": 1.03137994,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 2.054387093183573,
      "language_loss": 0.76728874,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.78990233,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.867799997329712
    },
    {
      "auxiliary_loss_clip": 0.01220467,
      "auxiliary_loss_mlp": 0.0103309,
      "balance_loss_clip": 0.94658178,
      "balance_loss_mlp": 1.02183664,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.6897684937449755,
      "language_loss": 0.7856701,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.80820572,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.7460241317749023
    },
    {
      "auxiliary_loss_clip": 0.01221938,
      "auxiliary_loss_mlp": 0.0104234,
      "balance_loss_clip": 1.02410865,
      "balance_loss_mlp": 1.03116405,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 1.8877897901560219,
      "language_loss": 0.74945271,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.77209544,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.8171281814575195
    },
    {
      "auxiliary_loss_clip": 0.01228527,
      "auxiliary_loss_mlp": 0.01040748,
      "balance_loss_clip": 1.06352544,
      "balance_loss_mlp": 1.02925634,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 2.715918773294313,
      "language_loss": 0.7601164,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.78280914,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.65242862701416
    },
    {
      "auxiliary_loss_clip": 0.01228478,
      "auxiliary_loss_mlp": 0.01042206,
      "balance_loss_clip": 0.98235989,
      "balance_loss_mlp": 1.03035641,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.3772532049114643,
      "language_loss": 0.80177915,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82448602,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.7141082286834717
    },
    {
      "auxiliary_loss_clip": 0.01225261,
      "auxiliary_loss_mlp": 0.01041203,
      "balance_loss_clip": 0.98730743,
      "balance_loss_mlp": 1.02994394,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 2.121794306112556,
      "language_loss": 0.7841965,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.80686116,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.7373268604278564
    },
    {
      "auxiliary_loss_clip": 0.01212889,
      "auxiliary_loss_mlp": 0.01037274,
      "balance_loss_clip": 0.86932743,
      "balance_loss_mlp": 1.02546,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 3.4994272276576,
      "language_loss": 0.78650463,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.80900621,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.8279058933258057
    },
    {
      "auxiliary_loss_clip": 0.01218055,
      "auxiliary_loss_mlp": 0.01038789,
      "balance_loss_clip": 0.94848144,
      "balance_loss_mlp": 1.02722609,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 2.752681843639125,
      "language_loss": 0.65765005,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.68021852,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 3.074650287628174
    },
    {
      "auxiliary_loss_clip": 0.01219863,
      "auxiliary_loss_mlp": 0.01041332,
      "balance_loss_clip": 0.98221755,
      "balance_loss_mlp": 1.02955365,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 1.817237260196046,
      "language_loss": 0.89066529,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.91327727,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.7645111083984375
    },
    {
      "auxiliary_loss_clip": 0.01231228,
      "auxiliary_loss_mlp": 0.01047467,
      "balance_loss_clip": 1.06487274,
      "balance_loss_mlp": 1.03633273,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 2.27031568017244,
      "language_loss": 0.79199505,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81478196,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 2.6137137413024902
    },
    {
      "auxiliary_loss_clip": 0.01201437,
      "auxiliary_loss_mlp": 0.01035253,
      "balance_loss_clip": 0.94237381,
      "balance_loss_mlp": 1.02354622,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 2.506627266568125,
      "language_loss": 0.79505426,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.8174212,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.730428457260132
    },
    {
      "auxiliary_loss_clip": 0.01216738,
      "auxiliary_loss_mlp": 0.01038987,
      "balance_loss_clip": 0.94786948,
      "balance_loss_mlp": 1.02683353,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.0378578552966524,
      "language_loss": 0.7416808,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.764238,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 2.70137095451355
    },
    {
      "auxiliary_loss_clip": 0.01227255,
      "auxiliary_loss_mlp": 0.01040964,
      "balance_loss_clip": 0.9553231,
      "balance_loss_mlp": 1.02898312,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 2.1848576675942284,
      "language_loss": 0.75638741,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.7790696,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.8804614543914795
    },
    {
      "auxiliary_loss_clip": 0.01218364,
      "auxiliary_loss_mlp": 0.01128898,
      "balance_loss_clip": 0.90889949,
      "balance_loss_mlp": 0.0,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 2.1041792855554995,
      "language_loss": 0.76847965,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.79195225,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.7544639110565186
    },
    {
      "auxiliary_loss_clip": 0.01231076,
      "auxiliary_loss_mlp": 0.01037181,
      "balance_loss_clip": 0.98805714,
      "balance_loss_mlp": 1.02483726,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.1791819915448682,
      "language_loss": 0.70759606,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73027861,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 3.5352225303649902
    },
    {
      "auxiliary_loss_clip": 0.01233319,
      "auxiliary_loss_mlp": 0.01040338,
      "balance_loss_clip": 1.02695441,
      "balance_loss_mlp": 1.02924585,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.53583119998272,
      "language_loss": 0.75946188,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78219843,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.689864158630371
    },
    {
      "auxiliary_loss_clip": 0.01220989,
      "auxiliary_loss_mlp": 0.01036434,
      "balance_loss_clip": 0.90561795,
      "balance_loss_mlp": 1.02469766,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 2.118422462325321,
      "language_loss": 0.80689096,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.82946515,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 3.720414161682129
    },
    {
      "auxiliary_loss_clip": 0.01229933,
      "auxiliary_loss_mlp": 0.01035684,
      "balance_loss_clip": 1.06411982,
      "balance_loss_mlp": 1.02465665,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 2.045077953350158,
      "language_loss": 0.78326654,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.80592275,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 4.557539939880371
    },
    {
      "auxiliary_loss_clip": 0.01219529,
      "auxiliary_loss_mlp": 0.01043765,
      "balance_loss_clip": 0.98709285,
      "balance_loss_mlp": 1.03095555,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 2.231543428808244,
      "language_loss": 0.75176919,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.77440214,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 2.6326398849487305
    },
    {
      "auxiliary_loss_clip": 0.01217546,
      "auxiliary_loss_mlp": 0.01038386,
      "balance_loss_clip": 0.90303314,
      "balance_loss_mlp": 1.02702558,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 1.7316942592492808,
      "language_loss": 0.80295044,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82550979,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.7854371070861816
    },
    {
      "auxiliary_loss_clip": 0.01220358,
      "auxiliary_loss_mlp": 0.0103994,
      "balance_loss_clip": 0.94329995,
      "balance_loss_mlp": 1.02821624,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 2.299699744556694,
      "language_loss": 0.89570725,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91831023,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.707085132598877
    },
    {
      "auxiliary_loss_clip": 0.01235816,
      "auxiliary_loss_mlp": 0.0104011,
      "balance_loss_clip": 1.0282402,
      "balance_loss_mlp": 1.02762294,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 2.416539396668676,
      "language_loss": 0.81670183,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.83946109,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 2.6493425369262695
    },
    {
      "auxiliary_loss_clip": 0.01233554,
      "auxiliary_loss_mlp": 0.01129281,
      "balance_loss_clip": 1.02543259,
      "balance_loss_mlp": 0.0,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 1.8682429357480648,
      "language_loss": 0.79823935,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.8218677,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.6955666542053223
    },
    {
      "auxiliary_loss_clip": 0.0123019,
      "auxiliary_loss_mlp": 0.01037932,
      "balance_loss_clip": 0.9533813,
      "balance_loss_mlp": 1.0264461,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 2.357822612477531,
      "language_loss": 0.76466846,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.7873497,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.7371978759765625
    },
    {
      "auxiliary_loss_clip": 0.01225403,
      "auxiliary_loss_mlp": 0.01037823,
      "balance_loss_clip": 1.02769852,
      "balance_loss_mlp": 1.02649248,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 6.286210375468758,
      "language_loss": 0.85118306,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.8738153,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.6852285861968994
    },
    {
      "auxiliary_loss_clip": 0.01230425,
      "auxiliary_loss_mlp": 0.01040418,
      "balance_loss_clip": 0.98619878,
      "balance_loss_mlp": 1.0292778,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.704646550315088,
      "language_loss": 0.81262863,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.8353371,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.7537972927093506
    },
    {
      "auxiliary_loss_clip": 0.01214536,
      "auxiliary_loss_mlp": 0.01039445,
      "balance_loss_clip": 0.90582955,
      "balance_loss_mlp": 1.02777433,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.3443059754551436,
      "language_loss": 0.77451235,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79705215,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.739086389541626
    },
    {
      "auxiliary_loss_clip": 0.01232368,
      "auxiliary_loss_mlp": 0.01040799,
      "balance_loss_clip": 1.02807379,
      "balance_loss_mlp": 1.02912211,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 2.1010668937760175,
      "language_loss": 0.88977802,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.91250974,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.6706650257110596
    },
    {
      "auxiliary_loss_clip": 0.01225575,
      "auxiliary_loss_mlp": 0.0103464,
      "balance_loss_clip": 0.94442844,
      "balance_loss_mlp": 1.02349973,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 2.035054298116174,
      "language_loss": 0.70640129,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.72900343,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.756871223449707
    },
    {
      "auxiliary_loss_clip": 0.01237704,
      "auxiliary_loss_mlp": 0.0112873,
      "balance_loss_clip": 1.03105152,
      "balance_loss_mlp": 0.0,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 2.2862021856351404,
      "language_loss": 0.88784933,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.91151369,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.725101947784424
    },
    {
      "auxiliary_loss_clip": 0.01221741,
      "auxiliary_loss_mlp": 0.01037553,
      "balance_loss_clip": 0.98455167,
      "balance_loss_mlp": 1.0260911,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.6613671361545137,
      "language_loss": 0.81607139,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.83866429,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.7512145042419434
    },
    {
      "auxiliary_loss_clip": 0.01217484,
      "auxiliary_loss_mlp": 0.01034515,
      "balance_loss_clip": 0.98568785,
      "balance_loss_mlp": 1.02423882,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 2.5285595580777667,
      "language_loss": 0.81493086,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83745086,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.652204990386963
    },
    {
      "auxiliary_loss_clip": 0.01231562,
      "auxiliary_loss_mlp": 0.01036923,
      "balance_loss_clip": 1.02722001,
      "balance_loss_mlp": 1.02567017,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 1.8855615330011266,
      "language_loss": 0.79680026,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81948507,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.6663315296173096
    },
    {
      "auxiliary_loss_clip": 0.01233958,
      "auxiliary_loss_mlp": 0.01039043,
      "balance_loss_clip": 1.06808662,
      "balance_loss_mlp": 1.0275569,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 1.8581088762393456,
      "language_loss": 0.79977584,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82250589,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.630394697189331
    },
    {
      "auxiliary_loss_clip": 0.01218795,
      "auxiliary_loss_mlp": 0.01036171,
      "balance_loss_clip": 0.98422718,
      "balance_loss_mlp": 1.02538872,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 1.8751365930115202,
      "language_loss": 0.69198823,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71453792,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 2.704052209854126
    },
    {
      "auxiliary_loss_clip": 0.01221559,
      "auxiliary_loss_mlp": 0.01040875,
      "balance_loss_clip": 0.90822381,
      "balance_loss_mlp": 1.02968764,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 2.7280718486740123,
      "language_loss": 0.95641643,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.97904086,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.8083596229553223
    },
    {
      "auxiliary_loss_clip": 0.01230577,
      "auxiliary_loss_mlp": 0.01042817,
      "balance_loss_clip": 1.06375027,
      "balance_loss_mlp": 1.03124762,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.2393904601539067,
      "language_loss": 0.82540971,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84814364,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 2.6178042888641357
    },
    {
      "auxiliary_loss_clip": 0.01226296,
      "auxiliary_loss_mlp": 0.01037095,
      "balance_loss_clip": 0.984519,
      "balance_loss_mlp": 1.02567494,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 1.7950551123579988,
      "language_loss": 0.67263007,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.69526398,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 2.8036296367645264
    },
    {
      "auxiliary_loss_clip": 0.01220681,
      "auxiliary_loss_mlp": 0.01029506,
      "balance_loss_clip": 0.86583364,
      "balance_loss_mlp": 1.01790142,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 1.7921115843550595,
      "language_loss": 0.72834235,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75084424,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.8475751876831055
    },
    {
      "auxiliary_loss_clip": 0.01221494,
      "auxiliary_loss_mlp": 0.01040535,
      "balance_loss_clip": 1.02714372,
      "balance_loss_mlp": 1.02951407,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.210854891588461,
      "language_loss": 0.78183746,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80445778,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.6793034076690674
    },
    {
      "auxiliary_loss_clip": 0.01207846,
      "auxiliary_loss_mlp": 0.01128647,
      "balance_loss_clip": 0.90884948,
      "balance_loss_mlp": 0.0,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 2.8884746556517,
      "language_loss": 0.79527694,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.8186419,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.8449747562408447
    },
    {
      "auxiliary_loss_clip": 0.01216088,
      "auxiliary_loss_mlp": 0.01037361,
      "balance_loss_clip": 1.02567339,
      "balance_loss_mlp": 1.02688837,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 2.0461708203069042,
      "language_loss": 0.89006048,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91259497,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 4.195743799209595
    },
    {
      "auxiliary_loss_clip": 0.01228395,
      "auxiliary_loss_mlp": 0.01038105,
      "balance_loss_clip": 1.06539297,
      "balance_loss_mlp": 1.02679801,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 2.1474464743697976,
      "language_loss": 0.8061462,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.82881117,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 4.6268956661224365
    },
    {
      "auxiliary_loss_clip": 0.01235574,
      "auxiliary_loss_mlp": 0.01037956,
      "balance_loss_clip": 1.02927566,
      "balance_loss_mlp": 1.02698278,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 2.39461893404025,
      "language_loss": 0.81743968,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.84017497,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 3.637411594390869
    },
    {
      "auxiliary_loss_clip": 0.01215259,
      "auxiliary_loss_mlp": 0.0103659,
      "balance_loss_clip": 0.90583152,
      "balance_loss_mlp": 1.02535498,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 1.9351663216069293,
      "language_loss": 0.64440048,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.66691893,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.7205147743225098
    },
    {
      "auxiliary_loss_clip": 0.01219739,
      "auxiliary_loss_mlp": 0.01039543,
      "balance_loss_clip": 0.94770646,
      "balance_loss_mlp": 1.02827144,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 1.74034097790216,
      "language_loss": 0.80793869,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.83053154,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.732818841934204
    },
    {
      "auxiliary_loss_clip": 0.01220994,
      "auxiliary_loss_mlp": 0.01038149,
      "balance_loss_clip": 0.94702733,
      "balance_loss_mlp": 1.0260551,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.6179295639716103,
      "language_loss": 0.82928193,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.85187334,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.7678544521331787
    },
    {
      "auxiliary_loss_clip": 0.01228151,
      "auxiliary_loss_mlp": 0.01040781,
      "balance_loss_clip": 1.06399822,
      "balance_loss_mlp": 1.02948546,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 1.7867260904675704,
      "language_loss": 0.51800686,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.5406962,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.7041444778442383
    },
    {
      "auxiliary_loss_clip": 0.01229415,
      "auxiliary_loss_mlp": 0.01039909,
      "balance_loss_clip": 0.9485153,
      "balance_loss_mlp": 1.02812481,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 2.677278656284657,
      "language_loss": 0.77878976,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80148298,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 2.7585999965667725
    },
    {
      "auxiliary_loss_clip": 0.01221439,
      "auxiliary_loss_mlp": 0.01038779,
      "balance_loss_clip": 0.98092818,
      "balance_loss_mlp": 1.02648854,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 2.9208686812257683,
      "language_loss": 0.72977352,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.7523756,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.7162647247314453
    },
    {
      "auxiliary_loss_clip": 0.01232301,
      "auxiliary_loss_mlp": 0.01037984,
      "balance_loss_clip": 1.06619275,
      "balance_loss_mlp": 1.0267489,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 1.9895682217711863,
      "language_loss": 0.82317996,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.84588277,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.615781545639038
    },
    {
      "auxiliary_loss_clip": 0.01218342,
      "auxiliary_loss_mlp": 0.01035588,
      "balance_loss_clip": 0.94447136,
      "balance_loss_mlp": 1.02481711,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 5.971599822808925,
      "language_loss": 0.69939768,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.72193694,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.7719292640686035
    },
    {
      "auxiliary_loss_clip": 0.01230337,
      "auxiliary_loss_mlp": 0.01038643,
      "balance_loss_clip": 1.06608236,
      "balance_loss_mlp": 1.0271275,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.3447716375762737,
      "language_loss": 0.91316921,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.93585902,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.6199679374694824
    },
    {
      "auxiliary_loss_clip": 0.01225716,
      "auxiliary_loss_mlp": 0.01043887,
      "balance_loss_clip": 1.02753067,
      "balance_loss_mlp": 1.03246665,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 1.7534524976328176,
      "language_loss": 0.88832396,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.91102004,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.706936836242676
    },
    {
      "auxiliary_loss_clip": 0.01221024,
      "auxiliary_loss_mlp": 0.01034088,
      "balance_loss_clip": 1.02365208,
      "balance_loss_mlp": 1.02377677,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 2.1689588020541546,
      "language_loss": 0.75971115,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.78226221,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.7112550735473633
    },
    {
      "auxiliary_loss_clip": 0.01223766,
      "auxiliary_loss_mlp": 0.01034203,
      "balance_loss_clip": 0.98689634,
      "balance_loss_mlp": 1.0229075,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 1.6220800416191106,
      "language_loss": 0.78795344,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81053317,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.7114012241363525
    },
    {
      "auxiliary_loss_clip": 0.01222387,
      "auxiliary_loss_mlp": 0.01037937,
      "balance_loss_clip": 1.02530015,
      "balance_loss_mlp": 1.02672553,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 2.7882434915283847,
      "language_loss": 0.85264969,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.8752529,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.6697916984558105
    },
    {
      "auxiliary_loss_clip": 0.01226517,
      "auxiliary_loss_mlp": 0.01036007,
      "balance_loss_clip": 0.98608047,
      "balance_loss_mlp": 1.02422905,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 2.1534884293726875,
      "language_loss": 0.71322078,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.73584604,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.7032151222229004
    },
    {
      "auxiliary_loss_clip": 0.01203168,
      "auxiliary_loss_mlp": 0.01034221,
      "balance_loss_clip": 0.82481003,
      "balance_loss_mlp": 1.0230931,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 1.852431311650141,
      "language_loss": 0.78618306,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.80855691,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.8204007148742676
    },
    {
      "auxiliary_loss_clip": 0.01202357,
      "auxiliary_loss_mlp": 0.01038597,
      "balance_loss_clip": 0.90733433,
      "balance_loss_mlp": 1.02678895,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.4255040437227815,
      "language_loss": 0.75279474,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.7752043,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.792429208755493
    },
    {
      "auxiliary_loss_clip": 0.01224751,
      "auxiliary_loss_mlp": 0.0103233,
      "balance_loss_clip": 1.02484465,
      "balance_loss_mlp": 1.0210228,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 1.7946128636897758,
      "language_loss": 0.71796429,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.74053514,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.779815673828125
    },
    {
      "auxiliary_loss_clip": 0.01213743,
      "auxiliary_loss_mlp": 0.01034859,
      "balance_loss_clip": 1.02129972,
      "balance_loss_mlp": 1.02395713,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 1.784453724171736,
      "language_loss": 0.73768032,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.76016635,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 2.672757387161255
    },
    {
      "auxiliary_loss_clip": 0.01194723,
      "auxiliary_loss_mlp": 0.01034697,
      "balance_loss_clip": 0.90174246,
      "balance_loss_mlp": 1.02347922,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 1.8437020244459885,
      "language_loss": 0.7655412,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.78783542,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 2.792264223098755
    },
    {
      "auxiliary_loss_clip": 0.01225784,
      "auxiliary_loss_mlp": 0.01127877,
      "balance_loss_clip": 1.06371236,
      "balance_loss_mlp": 0.0,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 2.2098838441357285,
      "language_loss": 0.84051883,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.86405545,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 2.6348891258239746
    },
    {
      "auxiliary_loss_clip": 0.01222839,
      "auxiliary_loss_mlp": 0.01036874,
      "balance_loss_clip": 0.98817956,
      "balance_loss_mlp": 1.02561474,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.7520972076350425,
      "language_loss": 0.85290021,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.87549734,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.7001538276672363
    },
    {
      "auxiliary_loss_clip": 0.01219467,
      "auxiliary_loss_mlp": 0.01032409,
      "balance_loss_clip": 0.94331264,
      "balance_loss_mlp": 1.02095258,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 3.0585664831176227,
      "language_loss": 0.72979134,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.75231004,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.756074905395508
    },
    {
      "auxiliary_loss_clip": 0.01138779,
      "auxiliary_loss_mlp": 0.01011749,
      "balance_loss_clip": 0.92872602,
      "balance_loss_mlp": 1.0059073,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.7202535512669153,
      "language_loss": 0.54538047,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56688571,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 4.4260172843933105
    },
    {
      "auxiliary_loss_clip": 0.0122542,
      "auxiliary_loss_mlp": 0.01036369,
      "balance_loss_clip": 1.06191266,
      "balance_loss_mlp": 1.02550316,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 1.9707257937779468,
      "language_loss": 0.70767105,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.73028898,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 2.6711957454681396
    },
    {
      "auxiliary_loss_clip": 0.01229447,
      "auxiliary_loss_mlp": 0.01043321,
      "balance_loss_clip": 1.06566226,
      "balance_loss_mlp": 1.03226995,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.3585248684011306,
      "language_loss": 0.78587008,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.8085978,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 3.6911520957946777
    },
    {
      "auxiliary_loss_clip": 0.01229984,
      "auxiliary_loss_mlp": 0.01039708,
      "balance_loss_clip": 1.06638813,
      "balance_loss_mlp": 1.02866316,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 2.2568898911176136,
      "language_loss": 0.73646909,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.759166,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 2.673736333847046
    },
    {
      "auxiliary_loss_clip": 0.01228886,
      "auxiliary_loss_mlp": 0.01037512,
      "balance_loss_clip": 0.95379275,
      "balance_loss_mlp": 1.02618074,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 1.8405770304588958,
      "language_loss": 0.73302931,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.75569332,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 3.742353916168213
    },
    {
      "auxiliary_loss_clip": 0.01231335,
      "auxiliary_loss_mlp": 0.01045106,
      "balance_loss_clip": 1.02560568,
      "balance_loss_mlp": 1.03417492,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 1.7654080656957252,
      "language_loss": 0.77314508,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79590952,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.702362060546875
    },
    {
      "auxiliary_loss_clip": 0.01215729,
      "auxiliary_loss_mlp": 0.01038508,
      "balance_loss_clip": 1.0228467,
      "balance_loss_mlp": 1.02660477,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 1.6970553362510066,
      "language_loss": 0.85473812,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.87728047,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.7397353649139404
    },
    {
      "auxiliary_loss_clip": 0.01213186,
      "auxiliary_loss_mlp": 0.01040805,
      "balance_loss_clip": 0.94730401,
      "balance_loss_mlp": 1.02975464,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 1.9512075051883726,
      "language_loss": 0.83742231,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85996222,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.826267719268799
    },
    {
      "auxiliary_loss_clip": 0.0121753,
      "auxiliary_loss_mlp": 0.01048258,
      "balance_loss_clip": 0.98619103,
      "balance_loss_mlp": 1.03711796,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 2.2818606550294853,
      "language_loss": 0.614048,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63670594,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.7371585369110107
    },
    {
      "auxiliary_loss_clip": 0.01218193,
      "auxiliary_loss_mlp": 0.01045704,
      "balance_loss_clip": 1.02613521,
      "balance_loss_mlp": 1.03424788,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 1.9952382278905942,
      "language_loss": 0.76936698,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79200602,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 2.668504238128662
    },
    {
      "auxiliary_loss_clip": 0.01215814,
      "auxiliary_loss_mlp": 0.01128324,
      "balance_loss_clip": 0.94396973,
      "balance_loss_mlp": 0.0,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 3.0434062501603134,
      "language_loss": 0.85365808,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.87709939,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.7284553050994873
    },
    {
      "auxiliary_loss_clip": 0.01207317,
      "auxiliary_loss_mlp": 0.01037258,
      "balance_loss_clip": 0.94384253,
      "balance_loss_mlp": 1.02631426,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 2.111284899635569,
      "language_loss": 0.76424515,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.78669089,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.791071653366089
    },
    {
      "auxiliary_loss_clip": 0.01222575,
      "auxiliary_loss_mlp": 0.0103741,
      "balance_loss_clip": 0.98951578,
      "balance_loss_mlp": 1.02645445,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 2.9078094058423845,
      "language_loss": 0.80209875,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82469857,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.9271836280822754
    },
    {
      "auxiliary_loss_clip": 0.01212423,
      "auxiliary_loss_mlp": 0.01042184,
      "balance_loss_clip": 0.9842965,
      "balance_loss_mlp": 1.03149724,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 3.2663523598154587,
      "language_loss": 0.84243882,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.86498493,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.749490261077881
    },
    {
      "auxiliary_loss_clip": 0.01217111,
      "auxiliary_loss_mlp": 0.01038357,
      "balance_loss_clip": 1.02084637,
      "balance_loss_mlp": 1.02716923,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 2.3483035269664394,
      "language_loss": 0.76644397,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.78899866,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.6836793422698975
    },
    {
      "auxiliary_loss_clip": 0.01221919,
      "auxiliary_loss_mlp": 0.01038813,
      "balance_loss_clip": 0.98787922,
      "balance_loss_mlp": 1.02766097,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 2.098118962784957,
      "language_loss": 0.81090659,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83351386,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.754650354385376
    },
    {
      "auxiliary_loss_clip": 0.0121902,
      "auxiliary_loss_mlp": 0.01036234,
      "balance_loss_clip": 0.98884904,
      "balance_loss_mlp": 1.02507651,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 2.710432932973262,
      "language_loss": 0.8703537,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.89290619,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.7317891120910645
    },
    {
      "auxiliary_loss_clip": 0.01227644,
      "auxiliary_loss_mlp": 0.01042028,
      "balance_loss_clip": 1.06531656,
      "balance_loss_mlp": 1.03154397,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 2.4854971920872915,
      "language_loss": 0.76320547,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.7859022,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.690920829772949
    },
    {
      "auxiliary_loss_clip": 0.01217917,
      "auxiliary_loss_mlp": 0.01038837,
      "balance_loss_clip": 0.95075369,
      "balance_loss_mlp": 1.02823901,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.240480053567059,
      "language_loss": 0.77040064,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79296815,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.740558385848999
    },
    {
      "auxiliary_loss_clip": 0.01212891,
      "auxiliary_loss_mlp": 0.01035604,
      "balance_loss_clip": 0.98117495,
      "balance_loss_mlp": 1.02523851,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.5710503899246953,
      "language_loss": 0.79526561,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.81775057,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.7487094402313232
    },
    {
      "auxiliary_loss_clip": 0.01221755,
      "auxiliary_loss_mlp": 0.01043751,
      "balance_loss_clip": 1.02650285,
      "balance_loss_mlp": 1.03295064,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 1.938107753585279,
      "language_loss": 0.69767058,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.72032559,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.774569272994995
    },
    {
      "auxiliary_loss_clip": 0.01140516,
      "auxiliary_loss_mlp": 0.01008263,
      "balance_loss_clip": 0.936023,
      "balance_loss_mlp": 1.00256526,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.9112968012753897,
      "language_loss": 0.58411658,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60560441,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 3.1135013103485107
    },
    {
      "auxiliary_loss_clip": 0.01215571,
      "auxiliary_loss_mlp": 0.01041252,
      "balance_loss_clip": 0.95039093,
      "balance_loss_mlp": 1.0295099,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 1.8367034896524665,
      "language_loss": 0.76970875,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.79227698,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.7697064876556396
    },
    {
      "auxiliary_loss_clip": 0.01211899,
      "auxiliary_loss_mlp": 0.01035895,
      "balance_loss_clip": 0.94728214,
      "balance_loss_mlp": 1.02496386,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 1.957510614776648,
      "language_loss": 0.85374248,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87622046,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 2.7621452808380127
    },
    {
      "auxiliary_loss_clip": 0.01219931,
      "auxiliary_loss_mlp": 0.01037348,
      "balance_loss_clip": 1.02416348,
      "balance_loss_mlp": 1.02697134,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 2.921797041479831,
      "language_loss": 0.80338567,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.82595849,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.6928422451019287
    },
    {
      "auxiliary_loss_clip": 0.01224113,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.02499008,
      "balance_loss_mlp": 1.0173775,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 2.3322649282989647,
      "language_loss": 0.82242924,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.84494787,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.711017608642578
    },
    {
      "auxiliary_loss_clip": 0.01231593,
      "auxiliary_loss_mlp": 0.01041761,
      "balance_loss_clip": 1.03171897,
      "balance_loss_mlp": 1.03081167,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 3.2537020774599155,
      "language_loss": 0.79359031,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81632388,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 3.5212948322296143
    },
    {
      "auxiliary_loss_clip": 0.01230405,
      "auxiliary_loss_mlp": 0.01037894,
      "balance_loss_clip": 0.98818511,
      "balance_loss_mlp": 1.02696228,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 2.2542766083424928,
      "language_loss": 0.78092664,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80360967,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 2.7557320594787598
    },
    {
      "auxiliary_loss_clip": 0.01223895,
      "auxiliary_loss_mlp": 0.01028839,
      "balance_loss_clip": 1.02618957,
      "balance_loss_mlp": 1.01862288,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 2.0264399985253956,
      "language_loss": 0.81203938,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83456677,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 3.8448901176452637
    },
    {
      "auxiliary_loss_clip": 0.01226555,
      "auxiliary_loss_mlp": 0.0103804,
      "balance_loss_clip": 1.0639751,
      "balance_loss_mlp": 1.02741826,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 2.9839739766009292,
      "language_loss": 0.94336784,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.96601379,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 2.658301591873169
    },
    {
      "auxiliary_loss_clip": 0.01227606,
      "auxiliary_loss_mlp": 0.01040145,
      "balance_loss_clip": 0.91294783,
      "balance_loss_mlp": 1.02900505,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 2.4176268444816555,
      "language_loss": 0.62211949,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64479697,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 3.780526876449585
    },
    {
      "auxiliary_loss_clip": 0.01135696,
      "auxiliary_loss_mlp": 0.01002673,
      "balance_loss_clip": 0.96589589,
      "balance_loss_mlp": 0.99687976,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8520675418860193,
      "language_loss": 0.68288875,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70427251,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.4521124362945557
    },
    {
      "auxiliary_loss_clip": 0.01223424,
      "auxiliary_loss_mlp": 0.01037711,
      "balance_loss_clip": 0.9496398,
      "balance_loss_mlp": 1.02689242,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.8839360497766553,
      "language_loss": 0.78635848,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.80896986,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 2.756791353225708
    },
    {
      "auxiliary_loss_clip": 0.01213616,
      "auxiliary_loss_mlp": 0.01041464,
      "balance_loss_clip": 0.98340386,
      "balance_loss_mlp": 1.03052616,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.5813885389699118,
      "language_loss": 0.8812288,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90377963,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.762028932571411
    },
    {
      "auxiliary_loss_clip": 0.01212532,
      "auxiliary_loss_mlp": 0.01037863,
      "balance_loss_clip": 0.98189807,
      "balance_loss_mlp": 1.02752185,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.37454077160248,
      "language_loss": 0.92360032,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.94610423,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.6714839935302734
    },
    {
      "auxiliary_loss_clip": 0.01208962,
      "auxiliary_loss_mlp": 0.01041358,
      "balance_loss_clip": 0.98458761,
      "balance_loss_mlp": 1.03051019,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 2.0958931360487303,
      "language_loss": 0.89229119,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91479439,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.7315618991851807
    },
    {
      "auxiliary_loss_clip": 0.01226795,
      "auxiliary_loss_mlp": 0.01127634,
      "balance_loss_clip": 1.06443048,
      "balance_loss_mlp": 0.0,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 2.322322046116175,
      "language_loss": 0.75328624,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.77683049,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 2.675062656402588
    },
    {
      "auxiliary_loss_clip": 0.01208265,
      "auxiliary_loss_mlp": 0.01047683,
      "balance_loss_clip": 0.98432487,
      "balance_loss_mlp": 1.03644156,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 2.550497468353177,
      "language_loss": 0.77671158,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.79927105,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.697268486022949
    },
    {
      "auxiliary_loss_clip": 0.01122957,
      "auxiliary_loss_mlp": 0.01005045,
      "balance_loss_clip": 0.95703793,
      "balance_loss_mlp": 0.99937099,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.8992457634718918,
      "language_loss": 0.56302243,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58430248,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.3395376205444336
    },
    {
      "auxiliary_loss_clip": 0.01227445,
      "auxiliary_loss_mlp": 0.01127435,
      "balance_loss_clip": 1.06630993,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 2.0435643369311527,
      "language_loss": 0.88899821,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.91254699,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.6390233039855957
    },
    {
      "auxiliary_loss_clip": 0.01228525,
      "auxiliary_loss_mlp": 0.01036062,
      "balance_loss_clip": 1.06410921,
      "balance_loss_mlp": 1.02556539,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 2.4569812001780384,
      "language_loss": 0.71038187,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73302776,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.8038716316223145
    },
    {
      "auxiliary_loss_clip": 0.01224152,
      "auxiliary_loss_mlp": 0.01039711,
      "balance_loss_clip": 0.99114114,
      "balance_loss_mlp": 1.02941775,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 1.7557905236740188,
      "language_loss": 0.76695698,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.7895956,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.740044355392456
    },
    {
      "auxiliary_loss_clip": 0.01219334,
      "auxiliary_loss_mlp": 0.01043097,
      "balance_loss_clip": 0.98405993,
      "balance_loss_mlp": 1.03223753,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 4.1016823539503795,
      "language_loss": 0.82265735,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.84528172,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 2.716252565383911
    },
    {
      "auxiliary_loss_clip": 0.01219009,
      "auxiliary_loss_mlp": 0.01033813,
      "balance_loss_clip": 0.98645955,
      "balance_loss_mlp": 1.02305412,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 2.8175195925449668,
      "language_loss": 0.80542082,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82794905,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.6236255168914795
    },
    {
      "auxiliary_loss_clip": 0.0121982,
      "auxiliary_loss_mlp": 0.01046938,
      "balance_loss_clip": 0.87028003,
      "balance_loss_mlp": 1.03521991,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 1.798502638426506,
      "language_loss": 0.70991135,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73257887,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 2.906285285949707
    },
    {
      "auxiliary_loss_clip": 0.01224745,
      "auxiliary_loss_mlp": 0.01039208,
      "balance_loss_clip": 1.02675486,
      "balance_loss_mlp": 1.0287118,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 2.0975260257895356,
      "language_loss": 0.77091873,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.79355824,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 2.6673362255096436
    },
    {
      "auxiliary_loss_clip": 0.01227673,
      "auxiliary_loss_mlp": 0.01033045,
      "balance_loss_clip": 1.06493688,
      "balance_loss_mlp": 1.02207136,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 1.776664923183026,
      "language_loss": 0.78621006,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.80881727,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.6764965057373047
    },
    {
      "auxiliary_loss_clip": 0.01224585,
      "auxiliary_loss_mlp": 0.0104221,
      "balance_loss_clip": 0.99047172,
      "balance_loss_mlp": 1.03187442,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 1.6837019310751036,
      "language_loss": 0.84370041,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86636841,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.8374216556549072
    },
    {
      "auxiliary_loss_clip": 0.01228146,
      "auxiliary_loss_mlp": 0.01037796,
      "balance_loss_clip": 1.06533611,
      "balance_loss_mlp": 1.02704978,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 2.5018179017248956,
      "language_loss": 0.74412251,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.76678187,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.6492671966552734
    },
    {
      "auxiliary_loss_clip": 0.01229225,
      "auxiliary_loss_mlp": 0.01034395,
      "balance_loss_clip": 1.02778816,
      "balance_loss_mlp": 1.02325475,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 1.9797017044431426,
      "language_loss": 0.81408584,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83672208,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 2.693225860595703
    },
    {
      "auxiliary_loss_clip": 0.01225123,
      "auxiliary_loss_mlp": 0.01037295,
      "balance_loss_clip": 0.94912809,
      "balance_loss_mlp": 1.0264585,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 1.7734286529062135,
      "language_loss": 0.88750601,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.91013026,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.736147880554199
    },
    {
      "auxiliary_loss_clip": 0.01229539,
      "auxiliary_loss_mlp": 0.01041691,
      "balance_loss_clip": 0.99400502,
      "balance_loss_mlp": 1.03093255,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 7.766669431293974,
      "language_loss": 0.83721411,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.8599264,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.742990732192993
    },
    {
      "auxiliary_loss_clip": 0.01223729,
      "auxiliary_loss_mlp": 0.01037629,
      "balance_loss_clip": 1.02866483,
      "balance_loss_mlp": 1.02767539,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 2.1553775885827746,
      "language_loss": 0.79986185,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.82247543,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 3.617685556411743
    },
    {
      "auxiliary_loss_clip": 0.01229307,
      "auxiliary_loss_mlp": 0.01039414,
      "balance_loss_clip": 1.06893492,
      "balance_loss_mlp": 1.02853632,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 2.0895104372728617,
      "language_loss": 0.73370445,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.75639164,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 2.6557905673980713
    },
    {
      "auxiliary_loss_clip": 0.01207045,
      "auxiliary_loss_mlp": 0.01035199,
      "balance_loss_clip": 0.98216736,
      "balance_loss_mlp": 1.02466071,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 1.8377164569775868,
      "language_loss": 0.69574606,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.7181685,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 3.658759355545044
    },
    {
      "auxiliary_loss_clip": 0.01225701,
      "auxiliary_loss_mlp": 0.01040243,
      "balance_loss_clip": 1.02540946,
      "balance_loss_mlp": 1.02945495,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 2.8639585916017336,
      "language_loss": 0.86749262,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.8901521,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 3.5319089889526367
    },
    {
      "auxiliary_loss_clip": 0.01221594,
      "auxiliary_loss_mlp": 0.01035502,
      "balance_loss_clip": 0.98632061,
      "balance_loss_mlp": 1.02514303,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 1.681372126761196,
      "language_loss": 0.79231596,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.81488687,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 3.6385819911956787
    },
    {
      "auxiliary_loss_clip": 0.01213839,
      "auxiliary_loss_mlp": 0.01037992,
      "balance_loss_clip": 0.90654558,
      "balance_loss_mlp": 1.02702427,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 2.0121434601356953,
      "language_loss": 0.87355089,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.89606923,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.7745559215545654
    },
    {
      "auxiliary_loss_clip": 0.01205478,
      "auxiliary_loss_mlp": 0.01043594,
      "balance_loss_clip": 0.86716104,
      "balance_loss_mlp": 1.03319907,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 1.9455614026348038,
      "language_loss": 0.69944549,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72193623,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.8470160961151123
    },
    {
      "auxiliary_loss_clip": 0.01228875,
      "auxiliary_loss_mlp": 0.01041765,
      "balance_loss_clip": 1.06506097,
      "balance_loss_mlp": 1.03136945,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 2.2205599918155774,
      "language_loss": 0.76634657,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.78905296,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.67893385887146
    },
    {
      "auxiliary_loss_clip": 0.0122253,
      "auxiliary_loss_mlp": 0.01039543,
      "balance_loss_clip": 0.94775534,
      "balance_loss_mlp": 1.02843308,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 1.8387313942729169,
      "language_loss": 0.69350815,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.71612883,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.748471260070801
    },
    {
      "auxiliary_loss_clip": 0.01219863,
      "auxiliary_loss_mlp": 0.01034348,
      "balance_loss_clip": 1.02232909,
      "balance_loss_mlp": 1.02412021,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 2.884316499328902,
      "language_loss": 0.83709502,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.85963714,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.692992925643921
    },
    {
      "auxiliary_loss_clip": 0.01224016,
      "auxiliary_loss_mlp": 0.01029904,
      "balance_loss_clip": 1.06363988,
      "balance_loss_mlp": 1.01888287,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.0953811280879386,
      "language_loss": 0.85594684,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.87848604,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 2.7124314308166504
    },
    {
      "auxiliary_loss_clip": 0.01223207,
      "auxiliary_loss_mlp": 0.01035138,
      "balance_loss_clip": 0.98820651,
      "balance_loss_mlp": 1.02492142,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 2.5864997352545025,
      "language_loss": 0.82441032,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.84699374,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.71551775932312
    },
    {
      "auxiliary_loss_clip": 0.01210308,
      "auxiliary_loss_mlp": 0.01036398,
      "balance_loss_clip": 0.98534179,
      "balance_loss_mlp": 1.02635503,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 1.681194681543702,
      "language_loss": 0.8802464,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90271354,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.841222047805786
    },
    {
      "auxiliary_loss_clip": 0.01221746,
      "auxiliary_loss_mlp": 0.01038799,
      "balance_loss_clip": 0.98788512,
      "balance_loss_mlp": 1.02886307,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 1.837032885615014,
      "language_loss": 0.78943896,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81204438,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.8021624088287354
    },
    {
      "auxiliary_loss_clip": 0.01219408,
      "auxiliary_loss_mlp": 0.01038313,
      "balance_loss_clip": 0.90867913,
      "balance_loss_mlp": 1.0276494,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 2.150768201457444,
      "language_loss": 0.83812207,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.86069924,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.859306573867798
    },
    {
      "auxiliary_loss_clip": 0.01226523,
      "auxiliary_loss_mlp": 0.01035513,
      "balance_loss_clip": 1.06400299,
      "balance_loss_mlp": 1.02437842,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.087771127961732,
      "language_loss": 0.6790216,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.70164198,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 2.7873575687408447
    },
    {
      "auxiliary_loss_clip": 0.01219608,
      "auxiliary_loss_mlp": 0.01036776,
      "balance_loss_clip": 0.98448402,
      "balance_loss_mlp": 1.02704298,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 2.0981778768712376,
      "language_loss": 0.74330831,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.76587212,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.756704092025757
    },
    {
      "auxiliary_loss_clip": 0.01221089,
      "auxiliary_loss_mlp": 0.01038015,
      "balance_loss_clip": 0.98519796,
      "balance_loss_mlp": 1.02701807,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 2.3718744279033497,
      "language_loss": 0.79132533,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81391633,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 2.673025131225586
    },
    {
      "auxiliary_loss_clip": 0.01222162,
      "auxiliary_loss_mlp": 0.01037837,
      "balance_loss_clip": 1.02343059,
      "balance_loss_mlp": 1.02697742,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 2.3027528130106094,
      "language_loss": 0.79067552,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.81327552,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.7187867164611816
    },
    {
      "auxiliary_loss_clip": 0.01210713,
      "auxiliary_loss_mlp": 0.01035264,
      "balance_loss_clip": 0.90602911,
      "balance_loss_mlp": 1.0247916,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 2.053386804575927,
      "language_loss": 0.84633803,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.86879784,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 2.715710163116455
    },
    {
      "auxiliary_loss_clip": 0.0114089,
      "auxiliary_loss_mlp": 0.01011475,
      "balance_loss_clip": 1.01007271,
      "balance_loss_mlp": 1.00613427,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.8416153502834665,
      "language_loss": 0.61684084,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63836449,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.229262351989746
    },
    {
      "auxiliary_loss_clip": 0.01138664,
      "auxiliary_loss_mlp": 0.01011608,
      "balance_loss_clip": 1.00827503,
      "balance_loss_mlp": 1.00605321,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.8918029131732983,
      "language_loss": 0.6243782,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64588094,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 3.2210707664489746
    },
    {
      "auxiliary_loss_clip": 0.01222835,
      "auxiliary_loss_mlp": 0.01035134,
      "balance_loss_clip": 0.94273901,
      "balance_loss_mlp": 1.0238626,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.0643910118475084,
      "language_loss": 0.77508569,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.79766536,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.7932238578796387
    },
    {
      "auxiliary_loss_clip": 0.01212443,
      "auxiliary_loss_mlp": 0.01127552,
      "balance_loss_clip": 0.98477125,
      "balance_loss_mlp": 0.0,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 1.944521141302535,
      "language_loss": 0.7881012,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.81150115,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.748434066772461
    },
    {
      "auxiliary_loss_clip": 0.01219857,
      "auxiliary_loss_mlp": 0.0103196,
      "balance_loss_clip": 0.98420703,
      "balance_loss_mlp": 1.02137387,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.092164516328352,
      "language_loss": 0.77020848,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.79272658,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.7824485301971436
    },
    {
      "auxiliary_loss_clip": 0.01222839,
      "auxiliary_loss_mlp": 0.01036535,
      "balance_loss_clip": 0.99023968,
      "balance_loss_mlp": 1.02585435,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 2.067056227069238,
      "language_loss": 0.77856642,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.8011601,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.7680647373199463
    },
    {
      "auxiliary_loss_clip": 0.0122498,
      "auxiliary_loss_mlp": 0.01038115,
      "balance_loss_clip": 0.95340276,
      "balance_loss_mlp": 1.02818537,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 1.9066896491091432,
      "language_loss": 0.91176915,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93440014,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 3.6416242122650146
    },
    {
      "auxiliary_loss_clip": 0.01223088,
      "auxiliary_loss_mlp": 0.01035425,
      "balance_loss_clip": 1.02316535,
      "balance_loss_mlp": 1.02516735,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 2.1432245970277415,
      "language_loss": 0.78536534,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.8079505,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 2.7379777431488037
    },
    {
      "auxiliary_loss_clip": 0.01225245,
      "auxiliary_loss_mlp": 0.01034919,
      "balance_loss_clip": 1.06324077,
      "balance_loss_mlp": 1.0246675,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 2.2705769708417174,
      "language_loss": 0.78939581,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.81199747,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 3.598726272583008
    },
    {
      "auxiliary_loss_clip": 0.01225769,
      "auxiliary_loss_mlp": 0.01030031,
      "balance_loss_clip": 0.98571759,
      "balance_loss_mlp": 1.01930773,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 1.8564345679787464,
      "language_loss": 0.8358475,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.85840547,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 3.6797492504119873
    },
    {
      "auxiliary_loss_clip": 0.01236223,
      "auxiliary_loss_mlp": 0.01039046,
      "balance_loss_clip": 1.02860975,
      "balance_loss_mlp": 1.02751231,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 1.7971457604788694,
      "language_loss": 0.861,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.8837527,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 3.5231099128723145
    },
    {
      "auxiliary_loss_clip": 0.01220904,
      "auxiliary_loss_mlp": 0.01034387,
      "balance_loss_clip": 1.0261941,
      "balance_loss_mlp": 1.02445138,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 2.1390322947868254,
      "language_loss": 0.7785362,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.80108905,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.6752405166625977
    },
    {
      "auxiliary_loss_clip": 0.01224038,
      "auxiliary_loss_mlp": 0.01047865,
      "balance_loss_clip": 1.0625844,
      "balance_loss_mlp": 1.03702331,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.535277736022648,
      "language_loss": 0.68395144,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70667052,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.6358935832977295
    },
    {
      "auxiliary_loss_clip": 0.01225524,
      "auxiliary_loss_mlp": 0.0104099,
      "balance_loss_clip": 1.02370477,
      "balance_loss_mlp": 1.03040457,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 1.8815318779950085,
      "language_loss": 0.77985179,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.80251694,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.7385568618774414
    },
    {
      "auxiliary_loss_clip": 0.01132525,
      "auxiliary_loss_mlp": 0.01007194,
      "balance_loss_clip": 0.9278121,
      "balance_loss_mlp": 1.00204372,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9276122320385036,
      "language_loss": 0.65700316,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.6784004,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.374321460723877
    },
    {
      "auxiliary_loss_clip": 0.01214825,
      "auxiliary_loss_mlp": 0.01032564,
      "balance_loss_clip": 0.83143914,
      "balance_loss_mlp": 1.0219965,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 1.9740856744441453,
      "language_loss": 0.83406961,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.85654342,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 2.8212292194366455
    },
    {
      "auxiliary_loss_clip": 0.0121913,
      "auxiliary_loss_mlp": 0.01032208,
      "balance_loss_clip": 0.98576272,
      "balance_loss_mlp": 1.0216397,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 2.4647538929290636,
      "language_loss": 0.79289162,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81540495,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 2.708639144897461
    },
    {
      "auxiliary_loss_clip": 0.01223893,
      "auxiliary_loss_mlp": 0.01037109,
      "balance_loss_clip": 1.02357006,
      "balance_loss_mlp": 1.02670217,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 2.1064421227557695,
      "language_loss": 0.79479039,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.8174004,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 2.677584648132324
    },
    {
      "auxiliary_loss_clip": 0.01211946,
      "auxiliary_loss_mlp": 0.01033979,
      "balance_loss_clip": 0.98457658,
      "balance_loss_mlp": 1.02300584,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 2.0285180806378733,
      "language_loss": 0.74936575,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.77182502,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.797846555709839
    },
    {
      "auxiliary_loss_clip": 0.0121676,
      "auxiliary_loss_mlp": 0.01035603,
      "balance_loss_clip": 1.02391648,
      "balance_loss_mlp": 1.02586973,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 2.455526695203486,
      "language_loss": 0.8048197,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.82734334,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.7418124675750732
    },
    {
      "auxiliary_loss_clip": 0.0122557,
      "auxiliary_loss_mlp": 0.01032638,
      "balance_loss_clip": 1.02744341,
      "balance_loss_mlp": 1.02227235,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 1.7730763965617695,
      "language_loss": 0.76601046,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.78859252,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.719839096069336
    },
    {
      "auxiliary_loss_clip": 0.01198895,
      "auxiliary_loss_mlp": 0.01042012,
      "balance_loss_clip": 0.90250993,
      "balance_loss_mlp": 1.03097951,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 3.361009358540471,
      "language_loss": 0.79573703,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81814611,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.752800226211548
    },
    {
      "auxiliary_loss_clip": 0.0123247,
      "auxiliary_loss_mlp": 0.01045398,
      "balance_loss_clip": 1.06925666,
      "balance_loss_mlp": 1.0346688,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 5.903064739348749,
      "language_loss": 0.67813778,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.70091641,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.670522451400757
    },
    {
      "auxiliary_loss_clip": 0.01223805,
      "auxiliary_loss_mlp": 0.01127861,
      "balance_loss_clip": 1.02540171,
      "balance_loss_mlp": 0.0,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 7.879289033474152,
      "language_loss": 0.82549846,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84901512,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.6995689868927
    },
    {
      "auxiliary_loss_clip": 0.01198215,
      "auxiliary_loss_mlp": 0.01030073,
      "balance_loss_clip": 0.90463096,
      "balance_loss_mlp": 1.01962459,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 1.8203823364589864,
      "language_loss": 0.78035343,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.80263633,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 2.9254112243652344
    },
    {
      "auxiliary_loss_clip": 0.01204596,
      "auxiliary_loss_mlp": 0.01034348,
      "balance_loss_clip": 0.90547287,
      "balance_loss_mlp": 1.02394676,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 2.1100716055866187,
      "language_loss": 0.87833846,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.90072787,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 2.840306282043457
    },
    {
      "auxiliary_loss_clip": 0.01193508,
      "auxiliary_loss_mlp": 0.01036431,
      "balance_loss_clip": 0.90589321,
      "balance_loss_mlp": 1.02577996,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 1.7474624998805541,
      "language_loss": 0.81544966,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.837749,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.7381224632263184
    },
    {
      "auxiliary_loss_clip": 0.01219422,
      "auxiliary_loss_mlp": 0.01038456,
      "balance_loss_clip": 1.02237535,
      "balance_loss_mlp": 1.02707791,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.1350964011453626,
      "language_loss": 0.77222621,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.79480499,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 2.701591968536377
    },
    {
      "auxiliary_loss_clip": 0.01222039,
      "auxiliary_loss_mlp": 0.01037585,
      "balance_loss_clip": 0.94652188,
      "balance_loss_mlp": 1.02719581,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 4.78103050179962,
      "language_loss": 0.84084523,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.86344147,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.695833683013916
    },
    {
      "auxiliary_loss_clip": 0.0121929,
      "auxiliary_loss_mlp": 0.01034887,
      "balance_loss_clip": 0.90387845,
      "balance_loss_mlp": 1.02377701,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 2.8207490491861744,
      "language_loss": 0.60272598,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62526774,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 2.9511208534240723
    },
    {
      "auxiliary_loss_clip": 0.01224152,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.02349448,
      "balance_loss_mlp": 1.02213728,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 3.090764806019778,
      "language_loss": 0.80618596,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.82875156,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.629279136657715
    },
    {
      "auxiliary_loss_clip": 0.01220906,
      "auxiliary_loss_mlp": 0.01127829,
      "balance_loss_clip": 0.98961568,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.4281657965126966,
      "language_loss": 0.86403108,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.88751841,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.746969223022461
    },
    {
      "auxiliary_loss_clip": 0.01213799,
      "auxiliary_loss_mlp": 0.01034059,
      "balance_loss_clip": 0.94824809,
      "balance_loss_mlp": 1.02363968,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.374435087705813,
      "language_loss": 0.84289992,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.86537856,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 3.6277964115142822
    },
    {
      "auxiliary_loss_clip": 0.01217878,
      "auxiliary_loss_mlp": 0.01028829,
      "balance_loss_clip": 0.9450388,
      "balance_loss_mlp": 1.01895273,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 2.8446955026426077,
      "language_loss": 0.60981584,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.63228291,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 2.8369107246398926
    },
    {
      "auxiliary_loss_clip": 0.01217447,
      "auxiliary_loss_mlp": 0.01036939,
      "balance_loss_clip": 0.98774302,
      "balance_loss_mlp": 1.02685964,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 1.9727491274171631,
      "language_loss": 0.77308643,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.79563034,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 3.6749515533447266
    },
    {
      "auxiliary_loss_clip": 0.01225283,
      "auxiliary_loss_mlp": 0.01038605,
      "balance_loss_clip": 1.06507099,
      "balance_loss_mlp": 1.02823961,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 1.9517310046485834,
      "language_loss": 0.75805938,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78069824,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 3.740875244140625
    },
    {
      "auxiliary_loss_clip": 0.01222039,
      "auxiliary_loss_mlp": 0.01032745,
      "balance_loss_clip": 1.06206155,
      "balance_loss_mlp": 1.02236211,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 1.8175477254344568,
      "language_loss": 0.80034226,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.82289016,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 3.491454601287842
    },
    {
      "auxiliary_loss_clip": 0.01202106,
      "auxiliary_loss_mlp": 0.01035981,
      "balance_loss_clip": 0.9456566,
      "balance_loss_mlp": 1.024912,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 2.015124481921483,
      "language_loss": 0.77568877,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.79806966,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.781083106994629
    },
    {
      "auxiliary_loss_clip": 0.01213533,
      "auxiliary_loss_mlp": 0.01037368,
      "balance_loss_clip": 0.94746405,
      "balance_loss_mlp": 1.02711558,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 2.056856600559266,
      "language_loss": 0.78717828,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.80968726,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.8924458026885986
    },
    {
      "auxiliary_loss_clip": 0.01222146,
      "auxiliary_loss_mlp": 0.01037282,
      "balance_loss_clip": 1.06398797,
      "balance_loss_mlp": 1.0274477,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.7636591031027273,
      "language_loss": 0.81827056,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.8408649,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.6560745239257812
    },
    {
      "auxiliary_loss_clip": 0.01216129,
      "auxiliary_loss_mlp": 0.01031452,
      "balance_loss_clip": 0.94886518,
      "balance_loss_mlp": 1.0216949,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 4.6161754787188265,
      "language_loss": 0.72690737,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.74938321,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.749913454055786
    },
    {
      "auxiliary_loss_clip": 0.01223792,
      "auxiliary_loss_mlp": 0.01034744,
      "balance_loss_clip": 1.02537692,
      "balance_loss_mlp": 1.02348423,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 3.6989506202724725,
      "language_loss": 0.82227379,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84485912,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.651806116104126
    },
    {
      "auxiliary_loss_clip": 0.01217142,
      "auxiliary_loss_mlp": 0.01035558,
      "balance_loss_clip": 0.98851281,
      "balance_loss_mlp": 1.02538371,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 2.0506986021142066,
      "language_loss": 0.72846377,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.75099075,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.7674171924591064
    },
    {
      "auxiliary_loss_clip": 0.01224455,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 0.9448123,
      "balance_loss_mlp": 1.02281654,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 13.431746479044929,
      "language_loss": 0.82695889,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.84953356,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 2.710502862930298
    },
    {
      "auxiliary_loss_clip": 0.01229523,
      "auxiliary_loss_mlp": 0.01038792,
      "balance_loss_clip": 0.98669916,
      "balance_loss_mlp": 1.02870655,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 2.1506485039769787,
      "language_loss": 0.83716607,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.85984921,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.7023630142211914
    },
    {
      "auxiliary_loss_clip": 0.01222846,
      "auxiliary_loss_mlp": 0.01035832,
      "balance_loss_clip": 0.98319006,
      "balance_loss_mlp": 1.02544272,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 2.137331603303107,
      "language_loss": 0.87098789,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89357471,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.9235737323760986
    },
    {
      "auxiliary_loss_clip": 0.01222182,
      "auxiliary_loss_mlp": 0.01040025,
      "balance_loss_clip": 0.98666543,
      "balance_loss_mlp": 1.0301125,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 1.6193729685143672,
      "language_loss": 0.85918391,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88180602,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.769636869430542
    },
    {
      "auxiliary_loss_clip": 0.01211185,
      "auxiliary_loss_mlp": 0.01030936,
      "balance_loss_clip": 0.94743443,
      "balance_loss_mlp": 1.02172756,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.70153445152724,
      "language_loss": 0.84565115,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86807233,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.708341121673584
    },
    {
      "auxiliary_loss_clip": 0.01134925,
      "auxiliary_loss_mlp": 0.01011428,
      "balance_loss_clip": 1.0064249,
      "balance_loss_mlp": 1.00603998,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.8587121476491945,
      "language_loss": 0.57965505,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60111862,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.2143850326538086
    },
    {
      "auxiliary_loss_clip": 0.01207599,
      "auxiliary_loss_mlp": 0.01035522,
      "balance_loss_clip": 0.98059571,
      "balance_loss_mlp": 1.02549028,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.0968955516078895,
      "language_loss": 0.85881841,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88124961,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 2.7912309169769287
    },
    {
      "auxiliary_loss_clip": 0.01205146,
      "auxiliary_loss_mlp": 0.01043489,
      "balance_loss_clip": 0.94369686,
      "balance_loss_mlp": 1.03249764,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 2.4180660605041977,
      "language_loss": 0.78692746,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.80941379,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 2.7149970531463623
    },
    {
      "auxiliary_loss_clip": 0.01123537,
      "auxiliary_loss_mlp": 0.0112276,
      "balance_loss_clip": 0.88474488,
      "balance_loss_mlp": 0.0,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.8550478581421991,
      "language_loss": 0.57954013,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.6020031,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.2714014053344727
    },
    {
      "auxiliary_loss_clip": 0.01222783,
      "auxiliary_loss_mlp": 0.01039336,
      "balance_loss_clip": 1.0270437,
      "balance_loss_mlp": 1.02873266,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 2.4211283437834448,
      "language_loss": 0.80510849,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.8277297,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.695596218109131
    },
    {
      "auxiliary_loss_clip": 0.01221849,
      "auxiliary_loss_mlp": 0.01032785,
      "balance_loss_clip": 1.02477026,
      "balance_loss_mlp": 1.02268159,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 2.775505922865807,
      "language_loss": 0.79015082,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.81269717,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.6744048595428467
    },
    {
      "auxiliary_loss_clip": 0.01220712,
      "auxiliary_loss_mlp": 0.01033391,
      "balance_loss_clip": 0.9456771,
      "balance_loss_mlp": 1.02260852,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 2.3107660383670887,
      "language_loss": 0.76163167,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.78417265,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.744274616241455
    },
    {
      "auxiliary_loss_clip": 0.0121873,
      "auxiliary_loss_mlp": 0.01035337,
      "balance_loss_clip": 1.05958641,
      "balance_loss_mlp": 1.02509737,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 2.057695970231504,
      "language_loss": 0.75304246,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77558309,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 2.6612963676452637
    },
    {
      "auxiliary_loss_clip": 0.01221915,
      "auxiliary_loss_mlp": 0.01037543,
      "balance_loss_clip": 1.02382648,
      "balance_loss_mlp": 1.02725577,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 2.171213910231569,
      "language_loss": 0.69185913,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71445376,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.8052754402160645
    },
    {
      "auxiliary_loss_clip": 0.01201665,
      "auxiliary_loss_mlp": 0.01035918,
      "balance_loss_clip": 0.90739989,
      "balance_loss_mlp": 1.02561867,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.4109842958315877,
      "language_loss": 0.89006567,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.91244149,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.750242233276367
    },
    {
      "auxiliary_loss_clip": 0.01225818,
      "auxiliary_loss_mlp": 0.0104128,
      "balance_loss_clip": 1.02313733,
      "balance_loss_mlp": 1.03058076,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 1.8543625112383113,
      "language_loss": 0.72908229,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.75175327,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 3.7008273601531982
    },
    {
      "auxiliary_loss_clip": 0.01210373,
      "auxiliary_loss_mlp": 0.01032158,
      "balance_loss_clip": 1.01915002,
      "balance_loss_mlp": 1.02169156,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 2.028217403949056,
      "language_loss": 0.78130054,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80372584,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 2.6800475120544434
    },
    {
      "auxiliary_loss_clip": 0.01225872,
      "auxiliary_loss_mlp": 0.01039986,
      "balance_loss_clip": 0.98548174,
      "balance_loss_mlp": 1.02899539,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.2519739709402056,
      "language_loss": 0.88363367,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90629232,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 3.6415069103240967
    },
    {
      "auxiliary_loss_clip": 0.01224132,
      "auxiliary_loss_mlp": 0.01037045,
      "balance_loss_clip": 0.98665273,
      "balance_loss_mlp": 1.02608418,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.050908469044794,
      "language_loss": 0.79641557,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.8190273,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 3.678212881088257
    },
    {
      "auxiliary_loss_clip": 0.01210788,
      "auxiliary_loss_mlp": 0.01035807,
      "balance_loss_clip": 0.90773666,
      "balance_loss_mlp": 1.02556729,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 1.9421131197856327,
      "language_loss": 0.75572681,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.7781927,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 3.739596128463745
    },
    {
      "auxiliary_loss_clip": 0.0121657,
      "auxiliary_loss_mlp": 0.0104007,
      "balance_loss_clip": 1.02227378,
      "balance_loss_mlp": 1.02948451,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 2.608871118558892,
      "language_loss": 0.76903439,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.79160082,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.7029147148132324
    },
    {
      "auxiliary_loss_clip": 0.01217324,
      "auxiliary_loss_mlp": 0.01041756,
      "balance_loss_clip": 1.02233553,
      "balance_loss_mlp": 1.03104472,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 2.307415233468777,
      "language_loss": 0.82724649,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.8498373,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.7295899391174316
    },
    {
      "auxiliary_loss_clip": 0.01221522,
      "auxiliary_loss_mlp": 0.01040395,
      "balance_loss_clip": 1.06383181,
      "balance_loss_mlp": 1.0296303,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 2.291934710473129,
      "language_loss": 0.85357302,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.87619215,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.5948657989501953
    },
    {
      "auxiliary_loss_clip": 0.0122651,
      "auxiliary_loss_mlp": 0.01127598,
      "balance_loss_clip": 1.06600046,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 2.7742974612690796,
      "language_loss": 0.69078183,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.71432287,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.694758653640747
    },
    {
      "auxiliary_loss_clip": 0.01219038,
      "auxiliary_loss_mlp": 0.01040976,
      "balance_loss_clip": 0.98398608,
      "balance_loss_mlp": 1.0308969,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.8380943301349253,
      "language_loss": 0.79336715,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81596726,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.6771438121795654
    },
    {
      "auxiliary_loss_clip": 0.01218725,
      "auxiliary_loss_mlp": 0.01037472,
      "balance_loss_clip": 0.8697536,
      "balance_loss_mlp": 1.02664208,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 1.7278366285822488,
      "language_loss": 0.94333994,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.96590191,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.801913022994995
    },
    {
      "auxiliary_loss_clip": 0.01224128,
      "auxiliary_loss_mlp": 0.01034823,
      "balance_loss_clip": 1.06434524,
      "balance_loss_mlp": 1.02402842,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 2.0034450701289854,
      "language_loss": 0.85261464,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87520421,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.667943239212036
    },
    {
      "auxiliary_loss_clip": 0.01219563,
      "auxiliary_loss_mlp": 0.01039678,
      "balance_loss_clip": 1.02371109,
      "balance_loss_mlp": 1.02874637,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 4.297521062962569,
      "language_loss": 0.86763382,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.89022619,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 2.6398961544036865
    },
    {
      "auxiliary_loss_clip": 0.01220657,
      "auxiliary_loss_mlp": 0.01040905,
      "balance_loss_clip": 0.98891854,
      "balance_loss_mlp": 1.0303843,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 1.9256641133426353,
      "language_loss": 0.7995162,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.82213175,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.7059648036956787
    },
    {
      "auxiliary_loss_clip": 0.01216411,
      "auxiliary_loss_mlp": 0.01041391,
      "balance_loss_clip": 0.9475283,
      "balance_loss_mlp": 1.03053069,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 4.0703422674244365,
      "language_loss": 0.78164411,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80422211,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.7452056407928467
    },
    {
      "auxiliary_loss_clip": 0.01204411,
      "auxiliary_loss_mlp": 0.01041766,
      "balance_loss_clip": 0.9838323,
      "balance_loss_mlp": 1.03107858,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 2.209952782336745,
      "language_loss": 0.93517995,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.95764172,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.636601448059082
    },
    {
      "auxiliary_loss_clip": 0.01209309,
      "auxiliary_loss_mlp": 0.01037692,
      "balance_loss_clip": 0.98242211,
      "balance_loss_mlp": 1.0273385,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 1.9231585889822476,
      "language_loss": 0.75474709,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.77721709,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.7341666221618652
    },
    {
      "auxiliary_loss_clip": 0.0121857,
      "auxiliary_loss_mlp": 0.01038006,
      "balance_loss_clip": 1.02215552,
      "balance_loss_mlp": 1.02740204,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 3.9084240995028092,
      "language_loss": 0.85848421,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.88104999,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 2.7682597637176514
    },
    {
      "auxiliary_loss_clip": 0.01206546,
      "auxiliary_loss_mlp": 0.01042536,
      "balance_loss_clip": 0.90833443,
      "balance_loss_mlp": 1.03198612,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 2.1444808899361485,
      "language_loss": 0.87641776,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.89890862,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 2.7732303142547607
    },
    {
      "auxiliary_loss_clip": 0.0122874,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 0.94862258,
      "balance_loss_mlp": 1.02081013,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 1.7992891085679048,
      "language_loss": 0.73527771,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.75788182,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 2.8225412368774414
    },
    {
      "auxiliary_loss_clip": 0.01218634,
      "auxiliary_loss_mlp": 0.01036464,
      "balance_loss_clip": 1.02361465,
      "balance_loss_mlp": 1.026021,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 2.4310200499743164,
      "language_loss": 0.85421932,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87677026,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.6583383083343506
    },
    {
      "auxiliary_loss_clip": 0.01126331,
      "auxiliary_loss_mlp": 0.01015513,
      "balance_loss_clip": 0.96285832,
      "balance_loss_mlp": 1.0103631,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.9909461366139642,
      "language_loss": 0.57303309,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59445155,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 3.344831943511963
    },
    {
      "auxiliary_loss_clip": 0.01198686,
      "auxiliary_loss_mlp": 0.01040554,
      "balance_loss_clip": 0.86771929,
      "balance_loss_mlp": 1.02977753,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 1.8473192654441326,
      "language_loss": 0.786807,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.80919945,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 2.803335189819336
    },
    {
      "auxiliary_loss_clip": 0.01212033,
      "auxiliary_loss_mlp": 0.01037814,
      "balance_loss_clip": 0.98308414,
      "balance_loss_mlp": 1.02718043,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.3670935228672816,
      "language_loss": 0.76448905,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.78698754,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 2.6632847785949707
    },
    {
      "auxiliary_loss_clip": 0.01201676,
      "auxiliary_loss_mlp": 0.0103927,
      "balance_loss_clip": 0.82738161,
      "balance_loss_mlp": 1.02826679,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 2.3320267741001253,
      "language_loss": 0.86128259,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88369203,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.7862725257873535
    },
    {
      "auxiliary_loss_clip": 0.01121973,
      "auxiliary_loss_mlp": 0.01007187,
      "balance_loss_clip": 0.88499671,
      "balance_loss_mlp": 1.00210881,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8545573675477448,
      "language_loss": 0.61614871,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63744032,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.2718756198883057
    },
    {
      "auxiliary_loss_clip": 0.01206402,
      "auxiliary_loss_mlp": 0.0103268,
      "balance_loss_clip": 0.9825958,
      "balance_loss_mlp": 1.02158153,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 2.3642428112350604,
      "language_loss": 0.86031842,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88270926,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 3.713339328765869
    },
    {
      "auxiliary_loss_clip": 0.01209987,
      "auxiliary_loss_mlp": 0.01040425,
      "balance_loss_clip": 0.90419471,
      "balance_loss_mlp": 1.03016675,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 3.2108658449268774,
      "language_loss": 0.84074688,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.86325103,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 2.7153382301330566
    },
    {
      "auxiliary_loss_clip": 0.01224677,
      "auxiliary_loss_mlp": 0.01031511,
      "balance_loss_clip": 0.98648793,
      "balance_loss_mlp": 1.02050841,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 2.0442332694466323,
      "language_loss": 0.82488304,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84744489,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 3.6648521423339844
    },
    {
      "auxiliary_loss_clip": 0.01216677,
      "auxiliary_loss_mlp": 0.01049779,
      "balance_loss_clip": 0.98561156,
      "balance_loss_mlp": 1.03961647,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 1.7884769459921845,
      "language_loss": 0.74111378,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76377833,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 3.6453049182891846
    },
    {
      "auxiliary_loss_clip": 0.01128056,
      "auxiliary_loss_mlp": 0.01008482,
      "balance_loss_clip": 0.96126199,
      "balance_loss_mlp": 1.0034039,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.972777796428356,
      "language_loss": 0.60459203,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62595737,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 4.173713207244873
    },
    {
      "auxiliary_loss_clip": 0.01208726,
      "auxiliary_loss_mlp": 0.01032877,
      "balance_loss_clip": 0.98211402,
      "balance_loss_mlp": 1.02236891,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 2.0356474811312006,
      "language_loss": 0.74591553,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76833153,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.856381416320801
    },
    {
      "auxiliary_loss_clip": 0.01219816,
      "auxiliary_loss_mlp": 0.01033047,
      "balance_loss_clip": 0.98818445,
      "balance_loss_mlp": 1.02220547,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 3.401740728883686,
      "language_loss": 0.72863472,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.75116336,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.848522424697876
    },
    {
      "auxiliary_loss_clip": 0.01225656,
      "auxiliary_loss_mlp": 0.01032505,
      "balance_loss_clip": 1.02341437,
      "balance_loss_mlp": 1.02171087,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 2.581439471851583,
      "language_loss": 0.78414327,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.80672491,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.707705497741699
    },
    {
      "auxiliary_loss_clip": 0.0122445,
      "auxiliary_loss_mlp": 0.01033457,
      "balance_loss_clip": 0.94973886,
      "balance_loss_mlp": 1.02343106,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 2.005931757532964,
      "language_loss": 0.74332821,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.76590723,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.799921989440918
    },
    {
      "auxiliary_loss_clip": 0.01116279,
      "auxiliary_loss_mlp": 0.01003929,
      "balance_loss_clip": 0.95831752,
      "balance_loss_mlp": 0.99875563,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8115755485826464,
      "language_loss": 0.60506344,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62626553,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.1865227222442627
    },
    {
      "auxiliary_loss_clip": 0.01222275,
      "auxiliary_loss_mlp": 0.01036889,
      "balance_loss_clip": 0.94818664,
      "balance_loss_mlp": 1.02627945,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 1.8030127759505026,
      "language_loss": 0.76635647,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.78894812,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.767364501953125
    },
    {
      "auxiliary_loss_clip": 0.01220793,
      "auxiliary_loss_mlp": 0.01035461,
      "balance_loss_clip": 0.98429388,
      "balance_loss_mlp": 1.02452397,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 2.5388170802986076,
      "language_loss": 0.87024307,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.8928057,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.733278512954712
    },
    {
      "auxiliary_loss_clip": 0.01221246,
      "auxiliary_loss_mlp": 0.01033883,
      "balance_loss_clip": 0.94555867,
      "balance_loss_mlp": 1.02380967,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 1.8278595318912754,
      "language_loss": 0.79109573,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81364703,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.7558624744415283
    },
    {
      "auxiliary_loss_clip": 0.01219186,
      "auxiliary_loss_mlp": 0.0103617,
      "balance_loss_clip": 1.02540112,
      "balance_loss_mlp": 1.02591181,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 1.8981047002262192,
      "language_loss": 0.73306584,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75561941,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.657149314880371
    },
    {
      "auxiliary_loss_clip": 0.01224269,
      "auxiliary_loss_mlp": 0.0103899,
      "balance_loss_clip": 0.98836672,
      "balance_loss_mlp": 1.0287199,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 2.3847042247888672,
      "language_loss": 0.8701182,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89275074,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.7091410160064697
    },
    {
      "auxiliary_loss_clip": 0.01216166,
      "auxiliary_loss_mlp": 0.01038213,
      "balance_loss_clip": 0.98177618,
      "balance_loss_mlp": 1.02765703,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.518745457907666,
      "language_loss": 0.73875356,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.76129735,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.7711570262908936
    },
    {
      "auxiliary_loss_clip": 0.01211077,
      "auxiliary_loss_mlp": 0.01038265,
      "balance_loss_clip": 0.98229444,
      "balance_loss_mlp": 1.02775085,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 2.83453674515599,
      "language_loss": 0.78923047,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.81172383,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.589765787124634
    },
    {
      "auxiliary_loss_clip": 0.01221109,
      "auxiliary_loss_mlp": 0.01040252,
      "balance_loss_clip": 0.9857372,
      "balance_loss_mlp": 1.02961898,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 1.9717751618982362,
      "language_loss": 0.90918422,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.9317978,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.7266182899475098
    },
    {
      "auxiliary_loss_clip": 0.01206428,
      "auxiliary_loss_mlp": 0.01037532,
      "balance_loss_clip": 0.94267261,
      "balance_loss_mlp": 1.02637959,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 1.6389829768711541,
      "language_loss": 0.90275908,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92519867,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 2.7746312618255615
    },
    {
      "auxiliary_loss_clip": 0.01224498,
      "auxiliary_loss_mlp": 0.01038902,
      "balance_loss_clip": 0.94926268,
      "balance_loss_mlp": 1.02848864,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 6.089273027398127,
      "language_loss": 0.79282737,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.81546134,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 2.6982057094573975
    },
    {
      "auxiliary_loss_clip": 0.01220841,
      "auxiliary_loss_mlp": 0.01038835,
      "balance_loss_clip": 0.98520827,
      "balance_loss_mlp": 1.0281961,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.2673084996019184,
      "language_loss": 0.92722619,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.94982302,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.762747287750244
    },
    {
      "auxiliary_loss_clip": 0.01222286,
      "auxiliary_loss_mlp": 0.01045073,
      "balance_loss_clip": 1.02228272,
      "balance_loss_mlp": 1.03364682,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 1.9123574614788597,
      "language_loss": 0.77886248,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.80153608,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 2.6424601078033447
    },
    {
      "auxiliary_loss_clip": 0.0121318,
      "auxiliary_loss_mlp": 0.01041312,
      "balance_loss_clip": 0.98850238,
      "balance_loss_mlp": 1.03026724,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 1.8696439255181956,
      "language_loss": 0.81485593,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83740085,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.766500234603882
    },
    {
      "auxiliary_loss_clip": 0.01216699,
      "auxiliary_loss_mlp": 0.01034317,
      "balance_loss_clip": 1.02381074,
      "balance_loss_mlp": 1.02295637,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 1.656926125087907,
      "language_loss": 0.81359828,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83610839,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 2.666642665863037
    },
    {
      "auxiliary_loss_clip": 0.01199875,
      "auxiliary_loss_mlp": 0.01038719,
      "balance_loss_clip": 0.86607128,
      "balance_loss_mlp": 1.02749515,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 1.6724694058429925,
      "language_loss": 0.84748423,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.86987013,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.811844825744629
    },
    {
      "auxiliary_loss_clip": 0.01215289,
      "auxiliary_loss_mlp": 0.01036485,
      "balance_loss_clip": 0.98350042,
      "balance_loss_mlp": 1.02577972,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 2.4390505660027353,
      "language_loss": 0.85320073,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.87571847,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.690946340560913
    },
    {
      "auxiliary_loss_clip": 0.01125418,
      "auxiliary_loss_mlp": 0.0100935,
      "balance_loss_clip": 0.99741006,
      "balance_loss_mlp": 1.00417614,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.7493874806177653,
      "language_loss": 0.61273354,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63408124,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 4.168447732925415
    },
    {
      "auxiliary_loss_clip": 0.01215832,
      "auxiliary_loss_mlp": 0.01036317,
      "balance_loss_clip": 0.95065504,
      "balance_loss_mlp": 1.02506328,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 2.163776052671387,
      "language_loss": 0.75904703,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.78156859,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 3.6047170162200928
    },
    {
      "auxiliary_loss_clip": 0.01207537,
      "auxiliary_loss_mlp": 0.01036789,
      "balance_loss_clip": 0.9045766,
      "balance_loss_mlp": 1.02551222,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 2.435210745239451,
      "language_loss": 0.74043196,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.7628752,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 2.764995813369751
    },
    {
      "auxiliary_loss_clip": 0.01124573,
      "auxiliary_loss_mlp": 0.010091,
      "balance_loss_clip": 0.95897496,
      "balance_loss_mlp": 1.00390291,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.8435068442142483,
      "language_loss": 0.64454317,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66587991,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 4.024532079696655
    },
    {
      "auxiliary_loss_clip": 0.01228284,
      "auxiliary_loss_mlp": 0.01035005,
      "balance_loss_clip": 1.064852,
      "balance_loss_mlp": 1.02388871,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 1.9483786665911607,
      "language_loss": 0.82760113,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.85023403,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.8423593044281006
    },
    {
      "auxiliary_loss_clip": 0.01205074,
      "auxiliary_loss_mlp": 0.01128094,
      "balance_loss_clip": 0.94813645,
      "balance_loss_mlp": 0.0,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.684988393343302,
      "language_loss": 0.79707336,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.82040513,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 3.6455302238464355
    },
    {
      "auxiliary_loss_clip": 0.01221701,
      "auxiliary_loss_mlp": 0.01127727,
      "balance_loss_clip": 0.98772115,
      "balance_loss_mlp": 0.0,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 6.063451271854564,
      "language_loss": 0.78298867,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.80648297,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.7101340293884277
    },
    {
      "auxiliary_loss_clip": 0.01218017,
      "auxiliary_loss_mlp": 0.01039252,
      "balance_loss_clip": 0.94781244,
      "balance_loss_mlp": 1.028476,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 2.1794306027642647,
      "language_loss": 0.75756538,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78013802,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.741417169570923
    },
    {
      "auxiliary_loss_clip": 0.01223747,
      "auxiliary_loss_mlp": 0.01029279,
      "balance_loss_clip": 1.02219892,
      "balance_loss_mlp": 1.01904488,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 2.641789613389145,
      "language_loss": 0.76260567,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.78513598,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.657919406890869
    },
    {
      "auxiliary_loss_clip": 0.01199879,
      "auxiliary_loss_mlp": 0.01044038,
      "balance_loss_clip": 0.86269486,
      "balance_loss_mlp": 1.03376842,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 1.5675321287082964,
      "language_loss": 0.74023485,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76267397,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.8819005489349365
    },
    {
      "auxiliary_loss_clip": 0.0120291,
      "auxiliary_loss_mlp": 0.01128041,
      "balance_loss_clip": 0.86661184,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.5471911324559835,
      "language_loss": 0.78336662,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.80667615,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.8810486793518066
    },
    {
      "auxiliary_loss_clip": 0.01128573,
      "auxiliary_loss_mlp": 0.01006444,
      "balance_loss_clip": 0.96050245,
      "balance_loss_mlp": 1.00134218,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.922196076926753,
      "language_loss": 0.5722599,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59361005,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.393996000289917
    },
    {
      "auxiliary_loss_clip": 0.01204927,
      "auxiliary_loss_mlp": 0.01127956,
      "balance_loss_clip": 0.9809674,
      "balance_loss_mlp": 0.0,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.5187579497116754,
      "language_loss": 0.78257018,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.80589902,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.7070086002349854
    },
    {
      "auxiliary_loss_clip": 0.01211748,
      "auxiliary_loss_mlp": 0.01034535,
      "balance_loss_clip": 0.94994211,
      "balance_loss_mlp": 1.0246172,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 1.8154132201027322,
      "language_loss": 0.9193961,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.94185901,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 2.7835798263549805
    },
    {
      "auxiliary_loss_clip": 0.01224719,
      "auxiliary_loss_mlp": 0.01034025,
      "balance_loss_clip": 1.0264982,
      "balance_loss_mlp": 1.02373171,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 1.9280385886488802,
      "language_loss": 0.84277248,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86535996,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.7083382606506348
    },
    {
      "auxiliary_loss_clip": 0.01208339,
      "auxiliary_loss_mlp": 0.01037236,
      "balance_loss_clip": 0.94322294,
      "balance_loss_mlp": 1.02678776,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 1.9611849746159749,
      "language_loss": 0.83724368,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.85969937,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 2.6506288051605225
    },
    {
      "auxiliary_loss_clip": 0.01212829,
      "auxiliary_loss_mlp": 0.01037325,
      "balance_loss_clip": 1.02031112,
      "balance_loss_mlp": 1.02681088,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 2.0210219295104497,
      "language_loss": 0.72385311,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74635464,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.737553358078003
    },
    {
      "auxiliary_loss_clip": 0.01218829,
      "auxiliary_loss_mlp": 0.01037998,
      "balance_loss_clip": 1.02464628,
      "balance_loss_mlp": 1.02705407,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 1.7850101548185335,
      "language_loss": 0.70118099,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72374928,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.6618640422821045
    },
    {
      "auxiliary_loss_clip": 0.0121175,
      "auxiliary_loss_mlp": 0.01042404,
      "balance_loss_clip": 0.98329961,
      "balance_loss_mlp": 1.03209805,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 2.025734596074511,
      "language_loss": 0.8353548,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.85789633,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 2.744520902633667
    },
    {
      "auxiliary_loss_clip": 0.01228105,
      "auxiliary_loss_mlp": 0.0103396,
      "balance_loss_clip": 1.02637529,
      "balance_loss_mlp": 1.02282,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.1629476297471286,
      "language_loss": 0.80601454,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.82863516,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 2.6197149753570557
    },
    {
      "auxiliary_loss_clip": 0.01202777,
      "auxiliary_loss_mlp": 0.01036796,
      "balance_loss_clip": 0.82720315,
      "balance_loss_mlp": 1.02596545,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 6.234507529824744,
      "language_loss": 0.70433998,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72673571,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.8268232345581055
    },
    {
      "auxiliary_loss_clip": 0.0120133,
      "auxiliary_loss_mlp": 0.01127914,
      "balance_loss_clip": 0.8269062,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 3.182505202271995,
      "language_loss": 0.75189531,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.77518773,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 2.8313329219818115
    },
    {
      "auxiliary_loss_clip": 0.01220504,
      "auxiliary_loss_mlp": 0.01038089,
      "balance_loss_clip": 1.02747416,
      "balance_loss_mlp": 1.02814746,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 1.9386641082320737,
      "language_loss": 0.77021253,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.7927984,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 2.747950315475464
    },
    {
      "auxiliary_loss_clip": 0.01212256,
      "auxiliary_loss_mlp": 0.01035521,
      "balance_loss_clip": 1.02182364,
      "balance_loss_mlp": 1.02490592,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 2.1409167951963566,
      "language_loss": 0.90850306,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.9309808,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 2.6950807571411133
    },
    {
      "auxiliary_loss_clip": 0.01217749,
      "auxiliary_loss_mlp": 0.01036771,
      "balance_loss_clip": 1.0220325,
      "balance_loss_mlp": 1.02636981,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 1.6570600015458514,
      "language_loss": 0.84252107,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86506635,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.622922420501709
    },
    {
      "auxiliary_loss_clip": 0.01222016,
      "auxiliary_loss_mlp": 0.01036507,
      "balance_loss_clip": 1.06387162,
      "balance_loss_mlp": 1.02603459,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.2396822885999836,
      "language_loss": 0.91599375,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.93857896,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.608158588409424
    },
    {
      "auxiliary_loss_clip": 0.01220546,
      "auxiliary_loss_mlp": 0.01039102,
      "balance_loss_clip": 0.95073748,
      "balance_loss_mlp": 1.02765799,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 1.82187756218373,
      "language_loss": 0.62381637,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.64641285,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 3.624063491821289
    },
    {
      "auxiliary_loss_clip": 0.01220924,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.02149785,
      "balance_loss_mlp": 1.01711655,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 2.4183351152026242,
      "language_loss": 0.73403728,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.75652504,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 3.6638669967651367
    },
    {
      "auxiliary_loss_clip": 0.01210691,
      "auxiliary_loss_mlp": 0.01032091,
      "balance_loss_clip": 0.98687822,
      "balance_loss_mlp": 1.02164781,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 2.141393541718933,
      "language_loss": 0.74623388,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.76866174,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 3.692715644836426
    },
    {
      "auxiliary_loss_clip": 0.01212488,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 0.86854529,
      "balance_loss_mlp": 1.01872301,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 2.5444820943914244,
      "language_loss": 0.72271729,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.74513531,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 2.88051176071167
    },
    {
      "auxiliary_loss_clip": 0.012275,
      "auxiliary_loss_mlp": 0.01037487,
      "balance_loss_clip": 1.0286243,
      "balance_loss_mlp": 1.02688301,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 2.1597985574944127,
      "language_loss": 0.78601623,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80866611,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.8607914447784424
    },
    {
      "auxiliary_loss_clip": 0.01214782,
      "auxiliary_loss_mlp": 0.01026688,
      "balance_loss_clip": 0.94346344,
      "balance_loss_mlp": 1.0172168,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 2.4997896100288504,
      "language_loss": 0.77412415,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.79653883,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 3.7889819145202637
    },
    {
      "auxiliary_loss_clip": 0.01215159,
      "auxiliary_loss_mlp": 0.01031215,
      "balance_loss_clip": 0.98487568,
      "balance_loss_mlp": 1.02103484,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 1.7647392842013072,
      "language_loss": 0.75782752,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.7802912,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.7788212299346924
    },
    {
      "auxiliary_loss_clip": 0.01218684,
      "auxiliary_loss_mlp": 0.01041245,
      "balance_loss_clip": 0.98603475,
      "balance_loss_mlp": 1.03115964,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 2.042591603370094,
      "language_loss": 0.71839797,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74099731,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.662247896194458
    },
    {
      "auxiliary_loss_clip": 0.01218759,
      "auxiliary_loss_mlp": 0.01030562,
      "balance_loss_clip": 0.98664665,
      "balance_loss_mlp": 1.02084661,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 2.3043381712355733,
      "language_loss": 0.82368886,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84618211,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.656966209411621
    },
    {
      "auxiliary_loss_clip": 0.01207399,
      "auxiliary_loss_mlp": 0.01042295,
      "balance_loss_clip": 0.94732285,
      "balance_loss_mlp": 1.03030825,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.2821468253103663,
      "language_loss": 0.87081897,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89331597,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.7156050205230713
    },
    {
      "auxiliary_loss_clip": 0.01196221,
      "auxiliary_loss_mlp": 0.01037346,
      "balance_loss_clip": 0.90134996,
      "balance_loss_mlp": 1.02640259,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.7184017421522266,
      "language_loss": 0.86494112,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.88727677,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.8349523544311523
    },
    {
      "auxiliary_loss_clip": 0.01225349,
      "auxiliary_loss_mlp": 0.01033421,
      "balance_loss_clip": 1.06711149,
      "balance_loss_mlp": 1.02315676,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 1.7744974478514453,
      "language_loss": 0.82631934,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84890711,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.6978836059570312
    },
    {
      "auxiliary_loss_clip": 0.01213352,
      "auxiliary_loss_mlp": 0.0103778,
      "balance_loss_clip": 0.90873563,
      "balance_loss_mlp": 1.02669978,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 2.9902941923067017,
      "language_loss": 0.87556595,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.89807725,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.824460983276367
    },
    {
      "auxiliary_loss_clip": 0.01221635,
      "auxiliary_loss_mlp": 0.01030774,
      "balance_loss_clip": 1.06203961,
      "balance_loss_mlp": 1.02033091,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 2.0220304787263768,
      "language_loss": 0.76763725,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.79016137,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.8290510177612305
    },
    {
      "auxiliary_loss_clip": 0.01214131,
      "auxiliary_loss_mlp": 0.01128465,
      "balance_loss_clip": 0.94805223,
      "balance_loss_mlp": 0.0,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 1.6617412079026828,
      "language_loss": 0.78082025,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.80424619,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 2.7791688442230225
    },
    {
      "auxiliary_loss_clip": 0.01118826,
      "auxiliary_loss_mlp": 0.01122396,
      "balance_loss_clip": 0.91908801,
      "balance_loss_mlp": 0.0,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7576224564787651,
      "language_loss": 0.60440403,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62681627,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.2971949577331543
    },
    {
      "auxiliary_loss_clip": 0.01213734,
      "auxiliary_loss_mlp": 0.01034407,
      "balance_loss_clip": 0.98425055,
      "balance_loss_mlp": 1.02424431,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.5322107024289464,
      "language_loss": 0.78712511,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.80960655,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 2.749469041824341
    },
    {
      "auxiliary_loss_clip": 0.01217668,
      "auxiliary_loss_mlp": 0.01038685,
      "balance_loss_clip": 0.94852853,
      "balance_loss_mlp": 1.02882028,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 3.124407925658383,
      "language_loss": 0.90783584,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.93039936,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.774329662322998
    },
    {
      "auxiliary_loss_clip": 0.01219686,
      "auxiliary_loss_mlp": 0.01031329,
      "balance_loss_clip": 0.90548348,
      "balance_loss_mlp": 1.02084446,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 2.5254341465295687,
      "language_loss": 0.77939105,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.80190122,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 2.8492259979248047
    },
    {
      "auxiliary_loss_clip": 0.01222898,
      "auxiliary_loss_mlp": 0.01043745,
      "balance_loss_clip": 1.0644803,
      "balance_loss_mlp": 1.03309333,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 2.9678536912915305,
      "language_loss": 0.796404,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81907046,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 2.6566336154937744
    },
    {
      "auxiliary_loss_clip": 0.01219114,
      "auxiliary_loss_mlp": 0.01039705,
      "balance_loss_clip": 1.0266279,
      "balance_loss_mlp": 1.02929842,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 2.142538751314335,
      "language_loss": 0.77039331,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.79298151,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.7761759757995605
    },
    {
      "auxiliary_loss_clip": 0.01201482,
      "auxiliary_loss_mlp": 0.01128023,
      "balance_loss_clip": 0.94589424,
      "balance_loss_mlp": 0.0,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 3.8473184576503874,
      "language_loss": 0.85560882,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.87890387,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 2.778399705886841
    },
    {
      "auxiliary_loss_clip": 0.01221891,
      "auxiliary_loss_mlp": 0.01034386,
      "balance_loss_clip": 1.06220233,
      "balance_loss_mlp": 1.02377677,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 2.5789876422672173,
      "language_loss": 0.82006335,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84262609,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.6679880619049072
    },
    {
      "auxiliary_loss_clip": 0.01219494,
      "auxiliary_loss_mlp": 0.01034099,
      "balance_loss_clip": 1.02414238,
      "balance_loss_mlp": 1.02392483,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 1.5244337459984934,
      "language_loss": 0.79772162,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.8202576,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 2.7203779220581055
    },
    {
      "auxiliary_loss_clip": 0.0121625,
      "auxiliary_loss_mlp": 0.01038764,
      "balance_loss_clip": 0.94502991,
      "balance_loss_mlp": 1.02851832,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 2.2436480895847395,
      "language_loss": 0.84869564,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.8712458,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.73563551902771
    },
    {
      "auxiliary_loss_clip": 0.01218996,
      "auxiliary_loss_mlp": 0.01028446,
      "balance_loss_clip": 0.94770455,
      "balance_loss_mlp": 1.01851022,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 1.6531666213439131,
      "language_loss": 0.83411413,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.8565886,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 2.752204656600952
    },
    {
      "auxiliary_loss_clip": 0.01219724,
      "auxiliary_loss_mlp": 0.01037418,
      "balance_loss_clip": 1.06512094,
      "balance_loss_mlp": 1.02775049,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.4146126949776843,
      "language_loss": 0.713732,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73630339,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 3.5319066047668457
    },
    {
      "auxiliary_loss_clip": 0.01212623,
      "auxiliary_loss_mlp": 0.01038996,
      "balance_loss_clip": 0.98743439,
      "balance_loss_mlp": 1.02776051,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 2.0287572124960245,
      "language_loss": 0.83587241,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.85838854,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 3.602477550506592
    },
    {
      "auxiliary_loss_clip": 0.01216062,
      "auxiliary_loss_mlp": 0.01036782,
      "balance_loss_clip": 0.98779124,
      "balance_loss_mlp": 1.02594018,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 2.1389975533140793,
      "language_loss": 0.87069046,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89321887,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 3.759874105453491
    },
    {
      "auxiliary_loss_clip": 0.01222847,
      "auxiliary_loss_mlp": 0.01036711,
      "balance_loss_clip": 1.02471817,
      "balance_loss_mlp": 1.02570772,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.2645807687792767,
      "language_loss": 0.71959507,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.74219066,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 2.6539809703826904
    },
    {
      "auxiliary_loss_clip": 0.01219323,
      "auxiliary_loss_mlp": 0.01033451,
      "balance_loss_clip": 1.02327919,
      "balance_loss_mlp": 1.02344942,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.54481417034873,
      "language_loss": 0.80520922,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82773703,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.7198872566223145
    },
    {
      "auxiliary_loss_clip": 0.01211671,
      "auxiliary_loss_mlp": 0.01127898,
      "balance_loss_clip": 0.90974414,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 1.889082049877589,
      "language_loss": 0.80910873,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.83250439,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 3.6977150440216064
    },
    {
      "auxiliary_loss_clip": 0.01218203,
      "auxiliary_loss_mlp": 0.01127607,
      "balance_loss_clip": 1.02392447,
      "balance_loss_mlp": 0.0,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 2.1760486949284017,
      "language_loss": 0.84702498,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.8704831,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.673722267150879
    },
    {
      "auxiliary_loss_clip": 0.01214988,
      "auxiliary_loss_mlp": 0.01034541,
      "balance_loss_clip": 0.98750347,
      "balance_loss_mlp": 1.02448535,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 1.927295314030911,
      "language_loss": 0.82589686,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.84839213,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.794018030166626
    },
    {
      "auxiliary_loss_clip": 0.01208323,
      "auxiliary_loss_mlp": 0.01127666,
      "balance_loss_clip": 0.94430906,
      "balance_loss_mlp": 0.0,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 1.8783223797083337,
      "language_loss": 0.72202337,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.74538326,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.784284830093384
    },
    {
      "auxiliary_loss_clip": 0.01204253,
      "auxiliary_loss_mlp": 0.01032975,
      "balance_loss_clip": 0.94344079,
      "balance_loss_mlp": 1.02289593,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 1.6983273067380809,
      "language_loss": 0.83785236,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.86022472,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.7300777435302734
    },
    {
      "auxiliary_loss_clip": 0.01216819,
      "auxiliary_loss_mlp": 0.01034929,
      "balance_loss_clip": 1.02358222,
      "balance_loss_mlp": 1.02424824,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 2.8350386461869705,
      "language_loss": 0.92944175,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.95195919,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.738039016723633
    },
    {
      "auxiliary_loss_clip": 0.01208721,
      "auxiliary_loss_mlp": 0.01035221,
      "balance_loss_clip": 0.98557377,
      "balance_loss_mlp": 1.0254395,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 1.6353581764857104,
      "language_loss": 0.7741363,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.79657578,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.8342690467834473
    },
    {
      "auxiliary_loss_clip": 0.01214963,
      "auxiliary_loss_mlp": 0.01031,
      "balance_loss_clip": 0.94436777,
      "balance_loss_mlp": 1.02080727,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 3.0974490953350196,
      "language_loss": 0.77104449,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.79350412,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.7460079193115234
    },
    {
      "auxiliary_loss_clip": 0.01214333,
      "auxiliary_loss_mlp": 0.01028466,
      "balance_loss_clip": 0.9849562,
      "balance_loss_mlp": 1.01823759,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 2.7979840775124964,
      "language_loss": 0.67291516,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.69534314,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.779834032058716
    },
    {
      "auxiliary_loss_clip": 0.01205764,
      "auxiliary_loss_mlp": 0.01040584,
      "balance_loss_clip": 0.94331551,
      "balance_loss_mlp": 1.03061771,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 1.7246207827664024,
      "language_loss": 0.82393903,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.84640247,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 2.7457573413848877
    },
    {
      "auxiliary_loss_clip": 0.01220637,
      "auxiliary_loss_mlp": 0.01040353,
      "balance_loss_clip": 1.06049097,
      "balance_loss_mlp": 1.03038144,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 2.3841829045591387,
      "language_loss": 0.82554406,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.84815395,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.6956474781036377
    },
    {
      "auxiliary_loss_clip": 0.01226342,
      "auxiliary_loss_mlp": 0.01040351,
      "balance_loss_clip": 0.94926965,
      "balance_loss_mlp": 1.02974176,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 1.9261394783916204,
      "language_loss": 0.8140378,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83670485,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.876417398452759
    },
    {
      "auxiliary_loss_clip": 0.0122067,
      "auxiliary_loss_mlp": 0.01033781,
      "balance_loss_clip": 1.02538466,
      "balance_loss_mlp": 1.02274776,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 2.4294604105987996,
      "language_loss": 0.76358151,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.78612602,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 2.6571226119995117
    },
    {
      "auxiliary_loss_clip": 0.01219485,
      "auxiliary_loss_mlp": 0.0103618,
      "balance_loss_clip": 0.98772341,
      "balance_loss_mlp": 1.0257436,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 3.1790011323167326,
      "language_loss": 0.80884731,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.83140397,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 2.6682708263397217
    },
    {
      "auxiliary_loss_clip": 0.01222229,
      "auxiliary_loss_mlp": 0.01036669,
      "balance_loss_clip": 1.06389761,
      "balance_loss_mlp": 1.02573144,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 2.152014543988575,
      "language_loss": 0.69401163,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.7166006,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.6273739337921143
    },
    {
      "auxiliary_loss_clip": 0.01218407,
      "auxiliary_loss_mlp": 0.01037762,
      "balance_loss_clip": 0.98667765,
      "balance_loss_mlp": 1.02771831,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 1.9165059547380705,
      "language_loss": 0.86712599,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.88968766,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 2.710939407348633
    },
    {
      "auxiliary_loss_clip": 0.01216884,
      "auxiliary_loss_mlp": 0.01043273,
      "balance_loss_clip": 1.024647,
      "balance_loss_mlp": 1.03247857,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 2.0886475283760486,
      "language_loss": 0.70736492,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.72996652,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 2.6785035133361816
    },
    {
      "auxiliary_loss_clip": 0.01223022,
      "auxiliary_loss_mlp": 0.01032844,
      "balance_loss_clip": 1.06514359,
      "balance_loss_mlp": 1.02209115,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 2.7201077938740874,
      "language_loss": 0.94225693,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96481556,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.6700634956359863
    },
    {
      "auxiliary_loss_clip": 0.01200206,
      "auxiliary_loss_mlp": 0.01036927,
      "balance_loss_clip": 0.90493542,
      "balance_loss_mlp": 1.02676427,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.3559992544666715,
      "language_loss": 0.7768079,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.7991792,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 2.810736894607544
    },
    {
      "auxiliary_loss_clip": 0.01219269,
      "auxiliary_loss_mlp": 0.01033429,
      "balance_loss_clip": 1.0250597,
      "balance_loss_mlp": 1.02265847,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.1009893451506994,
      "language_loss": 0.71282613,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.73535311,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.7478225231170654
    },
    {
      "auxiliary_loss_clip": 0.01213219,
      "auxiliary_loss_mlp": 0.0103003,
      "balance_loss_clip": 0.98444307,
      "balance_loss_mlp": 1.02005839,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.305382979878725,
      "language_loss": 0.86218929,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.88462174,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 2.6777350902557373
    },
    {
      "auxiliary_loss_clip": 0.01220403,
      "auxiliary_loss_mlp": 0.01032667,
      "balance_loss_clip": 0.90664172,
      "balance_loss_mlp": 1.02128839,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 1.860230027927688,
      "language_loss": 0.67009455,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.69262528,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 3.5879034996032715
    },
    {
      "auxiliary_loss_clip": 0.01220538,
      "auxiliary_loss_mlp": 0.01039269,
      "balance_loss_clip": 1.06533253,
      "balance_loss_mlp": 1.02905321,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 2.0752527369371,
      "language_loss": 0.75970608,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.78230417,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 3.6786701679229736
    },
    {
      "auxiliary_loss_clip": 0.01206265,
      "auxiliary_loss_mlp": 0.01033953,
      "balance_loss_clip": 0.90685749,
      "balance_loss_mlp": 1.02445769,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 2.1832674732785247,
      "language_loss": 0.78656918,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.80897135,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 3.731642961502075
    },
    {
      "auxiliary_loss_clip": 0.01216253,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.02263856,
      "balance_loss_mlp": 1.0197376,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.8241171638739615,
      "language_loss": 0.85300893,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87546706,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.6484055519104004
    },
    {
      "auxiliary_loss_clip": 0.01207859,
      "auxiliary_loss_mlp": 0.01047231,
      "balance_loss_clip": 0.86468494,
      "balance_loss_mlp": 1.03659141,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 2.1618356578835622,
      "language_loss": 0.8122046,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.83475554,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 3.7205262184143066
    },
    {
      "auxiliary_loss_clip": 0.01219942,
      "auxiliary_loss_mlp": 0.01037792,
      "balance_loss_clip": 1.02601445,
      "balance_loss_mlp": 1.02753353,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 1.7920971432023933,
      "language_loss": 0.85436749,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.8769449,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.7121846675872803
    },
    {
      "auxiliary_loss_clip": 0.01205169,
      "auxiliary_loss_mlp": 0.01037976,
      "balance_loss_clip": 0.98132461,
      "balance_loss_mlp": 1.02727652,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 5.398857913740429,
      "language_loss": 0.75745517,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.7798866,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.6762776374816895
    },
    {
      "auxiliary_loss_clip": 0.01218987,
      "auxiliary_loss_mlp": 0.01038238,
      "balance_loss_clip": 0.98176062,
      "balance_loss_mlp": 1.02823627,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 2.1382072857575682,
      "language_loss": 0.73757529,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76014757,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.75068736076355
    },
    {
      "auxiliary_loss_clip": 0.01205302,
      "auxiliary_loss_mlp": 0.01034327,
      "balance_loss_clip": 0.981408,
      "balance_loss_mlp": 1.02345514,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 1.890971823500587,
      "language_loss": 0.79951864,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82191497,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.7090158462524414
    },
    {
      "auxiliary_loss_clip": 0.01218585,
      "auxiliary_loss_mlp": 0.01127902,
      "balance_loss_clip": 0.98732406,
      "balance_loss_mlp": 0.0,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 1.7737442739080256,
      "language_loss": 0.79385036,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.81731522,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.783646821975708
    },
    {
      "auxiliary_loss_clip": 0.012083,
      "auxiliary_loss_mlp": 0.01034335,
      "balance_loss_clip": 1.02297735,
      "balance_loss_mlp": 1.02413654,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 2.1702068909519636,
      "language_loss": 0.71844858,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.74087489,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.655085325241089
    },
    {
      "auxiliary_loss_clip": 0.01214449,
      "auxiliary_loss_mlp": 0.01033976,
      "balance_loss_clip": 1.02107513,
      "balance_loss_mlp": 1.02382505,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 2.329209161753681,
      "language_loss": 0.79653001,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.81901425,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.6776275634765625
    },
    {
      "auxiliary_loss_clip": 0.01198061,
      "auxiliary_loss_mlp": 0.01035125,
      "balance_loss_clip": 0.94158709,
      "balance_loss_mlp": 1.02417541,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 1.6617418448933052,
      "language_loss": 0.79945695,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.82178885,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.800529718399048
    },
    {
      "auxiliary_loss_clip": 0.01221279,
      "auxiliary_loss_mlp": 0.0103978,
      "balance_loss_clip": 0.98680902,
      "balance_loss_mlp": 1.02883101,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 2.4624537014460643,
      "language_loss": 0.6498,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.67241061,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 2.648414134979248
    },
    {
      "auxiliary_loss_clip": 0.01210374,
      "auxiliary_loss_mlp": 0.01036892,
      "balance_loss_clip": 0.90437907,
      "balance_loss_mlp": 1.02679515,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 1.9424359339025765,
      "language_loss": 0.81954527,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.84201789,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.789442300796509
    },
    {
      "auxiliary_loss_clip": 0.01221007,
      "auxiliary_loss_mlp": 0.01035895,
      "balance_loss_clip": 0.95061028,
      "balance_loss_mlp": 1.02606583,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 2.0323252086964545,
      "language_loss": 0.86424118,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.88681018,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.7655465602874756
    },
    {
      "auxiliary_loss_clip": 0.01221582,
      "auxiliary_loss_mlp": 0.01042986,
      "balance_loss_clip": 1.02458549,
      "balance_loss_mlp": 1.03272223,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 2.039672552694647,
      "language_loss": 0.93300432,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.95564997,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.710444688796997
    },
    {
      "auxiliary_loss_clip": 0.01220343,
      "auxiliary_loss_mlp": 0.01034924,
      "balance_loss_clip": 1.02469039,
      "balance_loss_mlp": 1.02494597,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.212183269761317,
      "language_loss": 0.79305112,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.81560373,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 2.699882745742798
    },
    {
      "auxiliary_loss_clip": 0.01186107,
      "auxiliary_loss_mlp": 0.01036995,
      "balance_loss_clip": 0.82384115,
      "balance_loss_mlp": 1.02654004,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 1.8382504700151436,
      "language_loss": 0.80821085,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.83044183,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 2.832578420639038
    },
    {
      "auxiliary_loss_clip": 0.01188628,
      "auxiliary_loss_mlp": 0.01033523,
      "balance_loss_clip": 0.86363,
      "balance_loss_mlp": 1.02273488,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 1.6611097471659373,
      "language_loss": 0.73879969,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.76102126,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.8873798847198486
    },
    {
      "auxiliary_loss_clip": 0.01144969,
      "auxiliary_loss_mlp": 0.01010176,
      "balance_loss_clip": 0.89586425,
      "balance_loss_mlp": 1.00524068,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.0644198594144398,
      "language_loss": 0.7126888,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73424017,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 3.409972667694092
    },
    {
      "auxiliary_loss_clip": 0.01200612,
      "auxiliary_loss_mlp": 0.01038467,
      "balance_loss_clip": 0.94371009,
      "balance_loss_mlp": 1.02842975,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 2.108328730888849,
      "language_loss": 0.77604115,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79843193,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 2.83164644241333
    },
    {
      "auxiliary_loss_clip": 0.01201629,
      "auxiliary_loss_mlp": 0.01040725,
      "balance_loss_clip": 0.90715218,
      "balance_loss_mlp": 1.03053296,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 1.963803208136876,
      "language_loss": 0.78831077,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.81073427,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 2.727722644805908
    },
    {
      "auxiliary_loss_clip": 0.01133865,
      "auxiliary_loss_mlp": 0.01010142,
      "balance_loss_clip": 0.96823525,
      "balance_loss_mlp": 1.00520635,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7359323698069045,
      "language_loss": 0.5499022,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.57134229,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.374173402786255
    },
    {
      "auxiliary_loss_clip": 0.01196946,
      "auxiliary_loss_mlp": 0.01029568,
      "balance_loss_clip": 0.90351343,
      "balance_loss_mlp": 1.01955462,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.545931521578847,
      "language_loss": 0.76006627,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.78233135,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.75081205368042
    },
    {
      "auxiliary_loss_clip": 0.01211533,
      "auxiliary_loss_mlp": 0.01034243,
      "balance_loss_clip": 0.94344103,
      "balance_loss_mlp": 1.02372837,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 2.3072657879667866,
      "language_loss": 0.75265837,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77511609,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 2.759579658508301
    },
    {
      "auxiliary_loss_clip": 0.01219433,
      "auxiliary_loss_mlp": 0.01046732,
      "balance_loss_clip": 1.02442145,
      "balance_loss_mlp": 1.03571117,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.6708809838632823,
      "language_loss": 0.75873488,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.78139651,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 3.636859178543091
    },
    {
      "auxiliary_loss_clip": 0.01222416,
      "auxiliary_loss_mlp": 0.01040924,
      "balance_loss_clip": 0.99082327,
      "balance_loss_mlp": 1.0307498,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 2.0559620711163986,
      "language_loss": 0.74467558,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.76730901,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 3.7242753505706787
    },
    {
      "auxiliary_loss_clip": 0.01220972,
      "auxiliary_loss_mlp": 0.01041157,
      "balance_loss_clip": 1.02482367,
      "balance_loss_mlp": 1.03035057,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 2.0198504942688995,
      "language_loss": 0.76083392,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78345525,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 2.6876745223999023
    },
    {
      "auxiliary_loss_clip": 0.01221501,
      "auxiliary_loss_mlp": 0.01038272,
      "balance_loss_clip": 1.02338314,
      "balance_loss_mlp": 1.02738857,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 2.025263544288532,
      "language_loss": 0.77424574,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.79684353,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 3.5800082683563232
    },
    {
      "auxiliary_loss_clip": 0.01211644,
      "auxiliary_loss_mlp": 0.01036076,
      "balance_loss_clip": 0.98660672,
      "balance_loss_mlp": 1.02541256,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 1.8088065475212394,
      "language_loss": 0.79099119,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.8134684,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.8329617977142334
    },
    {
      "auxiliary_loss_clip": 0.01208848,
      "auxiliary_loss_mlp": 0.01037337,
      "balance_loss_clip": 0.90694928,
      "balance_loss_mlp": 1.0262742,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 1.9891266259640483,
      "language_loss": 0.8594169,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.88187873,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 3.6327967643737793
    },
    {
      "auxiliary_loss_clip": 0.01224514,
      "auxiliary_loss_mlp": 0.01127285,
      "balance_loss_clip": 1.02552783,
      "balance_loss_mlp": 0.0,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 1.5479800146136715,
      "language_loss": 0.77341223,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.79693025,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.7140421867370605
    },
    {
      "auxiliary_loss_clip": 0.01216368,
      "auxiliary_loss_mlp": 0.01037116,
      "balance_loss_clip": 1.02533913,
      "balance_loss_mlp": 1.02717972,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 1.9250166597814586,
      "language_loss": 0.82927108,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85180593,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.626667022705078
    },
    {
      "auxiliary_loss_clip": 0.01217281,
      "auxiliary_loss_mlp": 0.01036714,
      "balance_loss_clip": 1.06222486,
      "balance_loss_mlp": 1.02664053,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.10987416270139,
      "language_loss": 0.85145688,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87399685,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.6734514236450195
    },
    {
      "auxiliary_loss_clip": 0.01211166,
      "auxiliary_loss_mlp": 0.01042154,
      "balance_loss_clip": 0.98024577,
      "balance_loss_mlp": 1.03158629,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 1.8868428135734994,
      "language_loss": 0.79123682,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81377006,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.7062230110168457
    },
    {
      "auxiliary_loss_clip": 0.01211242,
      "auxiliary_loss_mlp": 0.01031196,
      "balance_loss_clip": 1.01952219,
      "balance_loss_mlp": 1.02072942,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 2.8640932664370657,
      "language_loss": 0.80304825,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82547265,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.703068256378174
    },
    {
      "auxiliary_loss_clip": 0.01218225,
      "auxiliary_loss_mlp": 0.01034375,
      "balance_loss_clip": 0.98532879,
      "balance_loss_mlp": 1.02423596,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 2.6951938847303505,
      "language_loss": 0.76308966,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.78561568,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.736959934234619
    },
    {
      "auxiliary_loss_clip": 0.01210297,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 0.90563422,
      "balance_loss_mlp": 1.01911032,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 2.135454942346143,
      "language_loss": 0.87292749,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.89532942,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.7780888080596924
    },
    {
      "auxiliary_loss_clip": 0.01217414,
      "auxiliary_loss_mlp": 0.01127508,
      "balance_loss_clip": 0.94966936,
      "balance_loss_mlp": 0.0,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 1.8472645453796077,
      "language_loss": 0.79353678,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.81698602,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.7123570442199707
    },
    {
      "auxiliary_loss_clip": 0.01202523,
      "auxiliary_loss_mlp": 0.0103444,
      "balance_loss_clip": 0.86860573,
      "balance_loss_mlp": 1.02370548,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 1.7593067800007083,
      "language_loss": 0.86422324,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.88659286,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.819626569747925
    },
    {
      "auxiliary_loss_clip": 0.01204047,
      "auxiliary_loss_mlp": 0.01039567,
      "balance_loss_clip": 0.98026466,
      "balance_loss_mlp": 1.02868342,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 3.173487031507285,
      "language_loss": 0.79810786,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.820544,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 2.8106539249420166
    },
    {
      "auxiliary_loss_clip": 0.01214049,
      "auxiliary_loss_mlp": 0.01037663,
      "balance_loss_clip": 0.9831059,
      "balance_loss_mlp": 1.02765524,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 1.7328059287382163,
      "language_loss": 0.74375832,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.76627553,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.783585786819458
    },
    {
      "auxiliary_loss_clip": 0.01221809,
      "auxiliary_loss_mlp": 0.01035086,
      "balance_loss_clip": 1.02595878,
      "balance_loss_mlp": 1.02493525,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 2.231002535193413,
      "language_loss": 0.86227626,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88484514,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.7130489349365234
    },
    {
      "auxiliary_loss_clip": 0.01214752,
      "auxiliary_loss_mlp": 0.01032151,
      "balance_loss_clip": 1.02414751,
      "balance_loss_mlp": 1.022012,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 6.369482799092971,
      "language_loss": 0.83400285,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.8564719,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 2.6708791255950928
    },
    {
      "auxiliary_loss_clip": 0.01223138,
      "auxiliary_loss_mlp": 0.01035161,
      "balance_loss_clip": 1.02375436,
      "balance_loss_mlp": 1.02468801,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 2.0866084457316796,
      "language_loss": 0.86403465,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.88661766,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.7325820922851562
    },
    {
      "auxiliary_loss_clip": 0.01202264,
      "auxiliary_loss_mlp": 0.01035151,
      "balance_loss_clip": 0.90626729,
      "balance_loss_mlp": 1.02492261,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 1.8597286526253525,
      "language_loss": 0.72945476,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.75182897,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.8614847660064697
    },
    {
      "auxiliary_loss_clip": 0.01220587,
      "auxiliary_loss_mlp": 0.01037063,
      "balance_loss_clip": 1.0629003,
      "balance_loss_mlp": 1.02638781,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 2.018248002313611,
      "language_loss": 0.75681788,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.77939439,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 2.67767596244812
    },
    {
      "auxiliary_loss_clip": 0.01214552,
      "auxiliary_loss_mlp": 0.01127572,
      "balance_loss_clip": 1.02336967,
      "balance_loss_mlp": 0.0,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 1.6064314307238143,
      "language_loss": 0.82879329,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.85221446,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 2.6493659019470215
    },
    {
      "auxiliary_loss_clip": 0.01224406,
      "auxiliary_loss_mlp": 0.01035986,
      "balance_loss_clip": 1.02609956,
      "balance_loss_mlp": 1.02543569,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 2.18833951014958,
      "language_loss": 0.73592836,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75853229,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 2.7876737117767334
    },
    {
      "auxiliary_loss_clip": 0.01215327,
      "auxiliary_loss_mlp": 0.01036897,
      "balance_loss_clip": 0.87253356,
      "balance_loss_mlp": 1.02640641,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 3.1850555238718683,
      "language_loss": 0.87829947,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90082169,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.7842180728912354
    },
    {
      "auxiliary_loss_clip": 0.0111442,
      "auxiliary_loss_mlp": 0.01008719,
      "balance_loss_clip": 0.99283266,
      "balance_loss_mlp": 1.00387907,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.7830619035928907,
      "language_loss": 0.5833143,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60454571,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 3.302994728088379
    },
    {
      "auxiliary_loss_clip": 0.01217942,
      "auxiliary_loss_mlp": 0.0104248,
      "balance_loss_clip": 0.98491818,
      "balance_loss_mlp": 1.03227556,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 2.046833695604353,
      "language_loss": 0.89071602,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.91332018,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 4.699731349945068
    },
    {
      "auxiliary_loss_clip": 0.01208642,
      "auxiliary_loss_mlp": 0.01036212,
      "balance_loss_clip": 0.98686051,
      "balance_loss_mlp": 1.02614462,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 2.3788115302750055,
      "language_loss": 0.7388972,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.76134574,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.6862566471099854
    },
    {
      "auxiliary_loss_clip": 0.01212863,
      "auxiliary_loss_mlp": 0.01040126,
      "balance_loss_clip": 0.94899619,
      "balance_loss_mlp": 1.02973127,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 1.8425025549378267,
      "language_loss": 0.78515363,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.80768347,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 3.7526538372039795
    },
    {
      "auxiliary_loss_clip": 0.01222635,
      "auxiliary_loss_mlp": 0.01127831,
      "balance_loss_clip": 1.06212163,
      "balance_loss_mlp": 0.0,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 1.7951252497130255,
      "language_loss": 0.83624107,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.85974574,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.8295798301696777
    },
    {
      "auxiliary_loss_clip": 0.01215466,
      "auxiliary_loss_mlp": 0.01031622,
      "balance_loss_clip": 1.02373791,
      "balance_loss_mlp": 1.02153134,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 1.9913846890883344,
      "language_loss": 0.84891421,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.8713851,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 3.665513753890991
    },
    {
      "auxiliary_loss_clip": 0.01220328,
      "auxiliary_loss_mlp": 0.01033883,
      "balance_loss_clip": 1.06335282,
      "balance_loss_mlp": 1.02361298,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 1.9691157060210922,
      "language_loss": 0.79103643,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.81357849,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.661255121231079
    },
    {
      "auxiliary_loss_clip": 0.0121341,
      "auxiliary_loss_mlp": 0.01126885,
      "balance_loss_clip": 1.02237988,
      "balance_loss_mlp": 0.0,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 1.7004663363114525,
      "language_loss": 0.79939914,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.82280219,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.8234400749206543
    },
    {
      "auxiliary_loss_clip": 0.01221604,
      "auxiliary_loss_mlp": 0.0103764,
      "balance_loss_clip": 0.95188689,
      "balance_loss_mlp": 1.02702498,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 1.8979668514158048,
      "language_loss": 0.82868385,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85127628,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.72611665725708
    },
    {
      "auxiliary_loss_clip": 0.01211609,
      "auxiliary_loss_mlp": 0.01036372,
      "balance_loss_clip": 0.94688326,
      "balance_loss_mlp": 1.02515435,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 1.8413413721715621,
      "language_loss": 0.87447125,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89695108,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 309.98071932792664
    },
    {
      "auxiliary_loss_clip": 0.01221425,
      "auxiliary_loss_mlp": 0.01037797,
      "balance_loss_clip": 0.94793344,
      "balance_loss_mlp": 1.02699089,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 3.0458244013967795,
      "language_loss": 0.71696347,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.73955572,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.802441120147705
    },
    {
      "auxiliary_loss_clip": 0.01206469,
      "auxiliary_loss_mlp": 0.01040798,
      "balance_loss_clip": 0.98398095,
      "balance_loss_mlp": 1.03000927,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 1.8977712242231253,
      "language_loss": 0.74865568,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.77112842,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 2.722126007080078
    },
    {
      "auxiliary_loss_clip": 0.0121763,
      "auxiliary_loss_mlp": 0.01038791,
      "balance_loss_clip": 0.98687661,
      "balance_loss_mlp": 1.02793717,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 1.7401498831800766,
      "language_loss": 0.77691972,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.7994839,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.714282274246216
    },
    {
      "auxiliary_loss_clip": 0.01206685,
      "auxiliary_loss_mlp": 0.01038794,
      "balance_loss_clip": 0.9049859,
      "balance_loss_mlp": 1.02791643,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 1.8420686130386008,
      "language_loss": 0.8534168,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.87587154,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.826768159866333
    },
    {
      "auxiliary_loss_clip": 0.01214471,
      "auxiliary_loss_mlp": 0.01036191,
      "balance_loss_clip": 1.02236032,
      "balance_loss_mlp": 1.02588511,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 1.9474920273378389,
      "language_loss": 0.78768837,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.81019497,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.666628837585449
    },
    {
      "auxiliary_loss_clip": 0.0122449,
      "auxiliary_loss_mlp": 0.01038796,
      "balance_loss_clip": 1.02585685,
      "balance_loss_mlp": 1.02845442,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 4.012510801447694,
      "language_loss": 0.81984222,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.84247506,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.729698896408081
    },
    {
      "auxiliary_loss_clip": 0.01208341,
      "auxiliary_loss_mlp": 0.01034527,
      "balance_loss_clip": 1.02395403,
      "balance_loss_mlp": 1.02360725,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 4.335556678248814,
      "language_loss": 0.73437572,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.75680441,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 2.6326162815093994
    },
    {
      "auxiliary_loss_clip": 0.01218112,
      "auxiliary_loss_mlp": 0.01127314,
      "balance_loss_clip": 0.98984492,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 2.0850362366161144,
      "language_loss": 0.79278982,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.81624413,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.752744197845459
    },
    {
      "auxiliary_loss_clip": 0.01116907,
      "auxiliary_loss_mlp": 0.01011872,
      "balance_loss_clip": 0.92631328,
      "balance_loss_mlp": 1.00693691,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.8995925716891555,
      "language_loss": 0.55949306,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.58078086,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.3057944774627686
    },
    {
      "auxiliary_loss_clip": 0.01215678,
      "auxiliary_loss_mlp": 0.01043311,
      "balance_loss_clip": 0.94711226,
      "balance_loss_mlp": 1.03286862,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 2.5290662503589973,
      "language_loss": 0.79244113,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.81503105,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 2.7952358722686768
    },
    {
      "auxiliary_loss_clip": 0.01214345,
      "auxiliary_loss_mlp": 0.01043365,
      "balance_loss_clip": 0.98538464,
      "balance_loss_mlp": 1.03312516,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 2.014915913377188,
      "language_loss": 0.69776225,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.72033936,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.7816250324249268
    },
    {
      "auxiliary_loss_clip": 0.01212528,
      "auxiliary_loss_mlp": 0.01040452,
      "balance_loss_clip": 0.98819268,
      "balance_loss_mlp": 1.02976537,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 1.8452610604239599,
      "language_loss": 0.76958162,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.7921114,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.749936103820801
    },
    {
      "auxiliary_loss_clip": 0.0112117,
      "auxiliary_loss_mlp": 0.0101472,
      "balance_loss_clip": 0.99499631,
      "balance_loss_mlp": 1.0098561,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.6814419407802931,
      "language_loss": 0.58330625,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60466516,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 3.341141939163208
    },
    {
      "auxiliary_loss_clip": 0.01201658,
      "auxiliary_loss_mlp": 0.01037886,
      "balance_loss_clip": 0.94382095,
      "balance_loss_mlp": 1.0280509,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 3.043779199998314,
      "language_loss": 0.76641291,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.78880835,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 2.709470748901367
    },
    {
      "auxiliary_loss_clip": 0.01205106,
      "auxiliary_loss_mlp": 0.01033932,
      "balance_loss_clip": 0.9442817,
      "balance_loss_mlp": 1.02330995,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 4.774287942594052,
      "language_loss": 0.75944042,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78183079,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.809690475463867
    },
    {
      "auxiliary_loss_clip": 0.0122261,
      "auxiliary_loss_mlp": 0.01040926,
      "balance_loss_clip": 0.988886,
      "balance_loss_mlp": 1.03078127,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 2.6882572284726995,
      "language_loss": 0.72948468,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.75212002,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 2.7733473777770996
    },
    {
      "auxiliary_loss_clip": 0.01227552,
      "auxiliary_loss_mlp": 0.01033866,
      "balance_loss_clip": 1.0660975,
      "balance_loss_mlp": 1.02283347,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 2.3561532520401913,
      "language_loss": 0.83847165,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.86108589,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 4.645304203033447
    },
    {
      "auxiliary_loss_clip": 0.01215974,
      "auxiliary_loss_mlp": 0.01034819,
      "balance_loss_clip": 0.98643607,
      "balance_loss_mlp": 1.02493644,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 1.7081314441715596,
      "language_loss": 0.79501951,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.81752741,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.6545932292938232
    },
    {
      "auxiliary_loss_clip": 0.01215148,
      "auxiliary_loss_mlp": 0.01044,
      "balance_loss_clip": 1.02573645,
      "balance_loss_mlp": 1.03415358,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 1.6472559873232713,
      "language_loss": 0.80737376,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.82996523,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.6448566913604736
    },
    {
      "auxiliary_loss_clip": 0.01222002,
      "auxiliary_loss_mlp": 0.0103892,
      "balance_loss_clip": 1.06264448,
      "balance_loss_mlp": 1.02845323,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 2.5896255248305464,
      "language_loss": 0.72964507,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75225431,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 3.6076552867889404
    },
    {
      "auxiliary_loss_clip": 0.01212371,
      "auxiliary_loss_mlp": 0.01049791,
      "balance_loss_clip": 0.98400652,
      "balance_loss_mlp": 1.03918767,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 1.6665974359256748,
      "language_loss": 0.6939612,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71658278,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 3.0103702545166016
    },
    {
      "auxiliary_loss_clip": 0.01221047,
      "auxiliary_loss_mlp": 0.01033163,
      "balance_loss_clip": 1.06412268,
      "balance_loss_mlp": 1.02271485,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 1.7135935841850911,
      "language_loss": 0.86528367,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.88782573,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 3.5564956665039062
    },
    {
      "auxiliary_loss_clip": 0.01225783,
      "auxiliary_loss_mlp": 0.01043083,
      "balance_loss_clip": 0.94855815,
      "balance_loss_mlp": 1.03250933,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 2.0952779691212644,
      "language_loss": 0.79508781,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.81777644,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.7345800399780273
    },
    {
      "auxiliary_loss_clip": 0.01219511,
      "auxiliary_loss_mlp": 0.01047305,
      "balance_loss_clip": 1.02531481,
      "balance_loss_mlp": 1.03640354,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 2.4194108712790694,
      "language_loss": 0.72085834,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.74352658,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.606532096862793
    },
    {
      "auxiliary_loss_clip": 0.01219376,
      "auxiliary_loss_mlp": 0.01034868,
      "balance_loss_clip": 1.02406991,
      "balance_loss_mlp": 1.02458024,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 1.8620699105483434,
      "language_loss": 0.88773251,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.91027492,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.77760648727417
    },
    {
      "auxiliary_loss_clip": 0.01206593,
      "auxiliary_loss_mlp": 0.01039236,
      "balance_loss_clip": 1.02289224,
      "balance_loss_mlp": 1.0293355,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 1.7696875759674184,
      "language_loss": 0.68835735,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71081567,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.766986846923828
    },
    {
      "auxiliary_loss_clip": 0.01220858,
      "auxiliary_loss_mlp": 0.01039226,
      "balance_loss_clip": 1.06234944,
      "balance_loss_mlp": 1.02871144,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 1.7694764400575156,
      "language_loss": 0.844028,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.86662889,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.7129931449890137
    },
    {
      "auxiliary_loss_clip": 0.01220285,
      "auxiliary_loss_mlp": 0.01044499,
      "balance_loss_clip": 1.06292474,
      "balance_loss_mlp": 1.03336525,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 2.18533659067518,
      "language_loss": 0.86953902,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.89218682,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.727339744567871
    },
    {
      "auxiliary_loss_clip": 0.01208302,
      "auxiliary_loss_mlp": 0.01033959,
      "balance_loss_clip": 0.94503069,
      "balance_loss_mlp": 1.02304542,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 1.9766120247276013,
      "language_loss": 0.76845288,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.79087555,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 2.727386713027954
    },
    {
      "auxiliary_loss_clip": 0.01213527,
      "auxiliary_loss_mlp": 0.01039945,
      "balance_loss_clip": 0.98587888,
      "balance_loss_mlp": 1.02962184,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 2.143055090464037,
      "language_loss": 0.7924794,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81501412,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.728583335876465
    },
    {
      "auxiliary_loss_clip": 0.0121274,
      "auxiliary_loss_mlp": 0.01038265,
      "balance_loss_clip": 0.90611517,
      "balance_loss_mlp": 1.02705359,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 1.7669067606044666,
      "language_loss": 0.82105911,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.84356916,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.804555892944336
    },
    {
      "auxiliary_loss_clip": 0.01220909,
      "auxiliary_loss_mlp": 0.01036949,
      "balance_loss_clip": 1.06479394,
      "balance_loss_mlp": 1.02707279,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 2.548621644786423,
      "language_loss": 0.66591787,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.68849647,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.6778564453125
    },
    {
      "auxiliary_loss_clip": 0.01217045,
      "auxiliary_loss_mlp": 0.01035325,
      "balance_loss_clip": 0.98287201,
      "balance_loss_mlp": 1.02499592,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.7190600317801201,
      "language_loss": 0.88623738,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.90876102,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 2.777348518371582
    },
    {
      "auxiliary_loss_clip": 0.01214484,
      "auxiliary_loss_mlp": 0.01040276,
      "balance_loss_clip": 1.02247047,
      "balance_loss_mlp": 1.02987504,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 3.2108477138013973,
      "language_loss": 0.76424813,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.78679574,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 2.688477039337158
    },
    {
      "auxiliary_loss_clip": 0.01214329,
      "auxiliary_loss_mlp": 0.01033799,
      "balance_loss_clip": 1.02435577,
      "balance_loss_mlp": 1.02402985,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 1.7264351764163937,
      "language_loss": 0.78613263,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.8086139,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.630478620529175
    },
    {
      "auxiliary_loss_clip": 0.01220353,
      "auxiliary_loss_mlp": 0.01034628,
      "balance_loss_clip": 1.02252841,
      "balance_loss_mlp": 1.02434576,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 2.413602337929441,
      "language_loss": 0.85314715,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.87569696,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 2.6611053943634033
    },
    {
      "auxiliary_loss_clip": 0.01129147,
      "auxiliary_loss_mlp": 0.01013729,
      "balance_loss_clip": 1.03781343,
      "balance_loss_mlp": 1.00919867,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.0613556209440447,
      "language_loss": 0.66328657,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.68471527,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.3066306114196777
    },
    {
      "auxiliary_loss_clip": 0.01206693,
      "auxiliary_loss_mlp": 0.01038599,
      "balance_loss_clip": 0.94331729,
      "balance_loss_mlp": 1.0284667,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 2.073832098726894,
      "language_loss": 0.69242918,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.71488214,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 2.8850784301757812
    },
    {
      "auxiliary_loss_clip": 0.01136433,
      "auxiliary_loss_mlp": 0.01122573,
      "balance_loss_clip": 0.92414224,
      "balance_loss_mlp": 0.0,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.9331244989692222,
      "language_loss": 0.61150873,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.63409877,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.41064453125
    },
    {
      "auxiliary_loss_clip": 0.01221945,
      "auxiliary_loss_mlp": 0.0103574,
      "balance_loss_clip": 1.06247199,
      "balance_loss_mlp": 1.02502298,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 1.788983858598234,
      "language_loss": 0.77498543,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.79756224,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 2.619774103164673
    },
    {
      "auxiliary_loss_clip": 0.01126147,
      "auxiliary_loss_mlp": 0.01013565,
      "balance_loss_clip": 1.03541517,
      "balance_loss_mlp": 1.00901115,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 0.9919307198988149,
      "language_loss": 0.64039624,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66179335,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 3.0790059566497803
    },
    {
      "auxiliary_loss_clip": 0.01218351,
      "auxiliary_loss_mlp": 0.01035059,
      "balance_loss_clip": 1.02443731,
      "balance_loss_mlp": 1.02495027,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 1.828463176911667,
      "language_loss": 0.80800867,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.8305428,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 3.653841733932495
    },
    {
      "auxiliary_loss_clip": 0.01208212,
      "auxiliary_loss_mlp": 0.01034811,
      "balance_loss_clip": 0.82945806,
      "balance_loss_mlp": 1.02453518,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 1.9184740283818693,
      "language_loss": 0.82784164,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.85027182,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 2.8750734329223633
    },
    {
      "auxiliary_loss_clip": 0.01218356,
      "auxiliary_loss_mlp": 0.01036229,
      "balance_loss_clip": 1.02469122,
      "balance_loss_mlp": 1.02511287,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 1.7389932629839662,
      "language_loss": 0.86297047,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88551629,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 3.6260204315185547
    },
    {
      "auxiliary_loss_clip": 0.01216427,
      "auxiliary_loss_mlp": 0.01127046,
      "balance_loss_clip": 0.98997247,
      "balance_loss_mlp": 0.0,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 2.00633778748332,
      "language_loss": 0.80467808,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.82811284,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.7922816276550293
    },
    {
      "auxiliary_loss_clip": 0.01221654,
      "auxiliary_loss_mlp": 0.01039555,
      "balance_loss_clip": 1.02736425,
      "balance_loss_mlp": 1.0291121,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 2.737143737754973,
      "language_loss": 0.76073492,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78334701,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 3.6910741329193115
    },
    {
      "auxiliary_loss_clip": 0.01203809,
      "auxiliary_loss_mlp": 0.01040213,
      "balance_loss_clip": 1.0204258,
      "balance_loss_mlp": 1.03026557,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.777075650989988,
      "language_loss": 0.86120892,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.88364917,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.740769624710083
    },
    {
      "auxiliary_loss_clip": 0.01202833,
      "auxiliary_loss_mlp": 0.01034142,
      "balance_loss_clip": 0.98109704,
      "balance_loss_mlp": 1.02348411,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 1.9347119139976137,
      "language_loss": 0.74787927,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.77024901,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.690810203552246
    },
    {
      "auxiliary_loss_clip": 0.01216901,
      "auxiliary_loss_mlp": 0.01035109,
      "balance_loss_clip": 1.02285051,
      "balance_loss_mlp": 1.02498257,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 2.3008225615212083,
      "language_loss": 0.74935734,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.77187747,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 3.578345537185669
    },
    {
      "auxiliary_loss_clip": 0.01199011,
      "auxiliary_loss_mlp": 0.01039894,
      "balance_loss_clip": 0.9420296,
      "balance_loss_mlp": 1.03007734,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.8199982986428729,
      "language_loss": 0.73955703,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76194608,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.681640625
    },
    {
      "auxiliary_loss_clip": 0.01222163,
      "auxiliary_loss_mlp": 0.01037802,
      "balance_loss_clip": 1.06303692,
      "balance_loss_mlp": 1.02712131,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 1.8576930706329995,
      "language_loss": 0.82838738,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85098702,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.6935675144195557
    },
    {
      "auxiliary_loss_clip": 0.01217768,
      "auxiliary_loss_mlp": 0.01037356,
      "balance_loss_clip": 0.98581839,
      "balance_loss_mlp": 1.02666938,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 2.181474115386808,
      "language_loss": 0.72399688,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.74654812,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.7067513465881348
    },
    {
      "auxiliary_loss_clip": 0.01201151,
      "auxiliary_loss_mlp": 0.01033637,
      "balance_loss_clip": 0.87131017,
      "balance_loss_mlp": 1.02333713,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 2.037162001274475,
      "language_loss": 0.71188414,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.73423207,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.806422710418701
    },
    {
      "auxiliary_loss_clip": 0.01214175,
      "auxiliary_loss_mlp": 0.01046706,
      "balance_loss_clip": 0.94555652,
      "balance_loss_mlp": 1.03622794,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 1.9224446301699007,
      "language_loss": 0.69252884,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.7151376,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.816861629486084
    },
    {
      "auxiliary_loss_clip": 0.01215236,
      "auxiliary_loss_mlp": 0.01039091,
      "balance_loss_clip": 1.02629328,
      "balance_loss_mlp": 1.02969146,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 1.892119057763836,
      "language_loss": 0.86131167,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88385493,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 2.729114294052124
    },
    {
      "auxiliary_loss_clip": 0.01126898,
      "auxiliary_loss_mlp": 0.01010709,
      "balance_loss_clip": 0.95904505,
      "balance_loss_mlp": 1.00574958,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.0832176281484136,
      "language_loss": 0.62481499,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.646191,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 3.099411725997925
    },
    {
      "auxiliary_loss_clip": 0.0122996,
      "auxiliary_loss_mlp": 0.01045528,
      "balance_loss_clip": 1.06631672,
      "balance_loss_mlp": 1.0349896,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 1.807849379628114,
      "language_loss": 0.86547744,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.88823235,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.6191093921661377
    },
    {
      "auxiliary_loss_clip": 0.01219456,
      "auxiliary_loss_mlp": 0.01043355,
      "balance_loss_clip": 0.91100466,
      "balance_loss_mlp": 1.03296614,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 1.9036678104202436,
      "language_loss": 0.78493291,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80756104,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.800062417984009
    },
    {
      "auxiliary_loss_clip": 0.01198933,
      "auxiliary_loss_mlp": 0.01039599,
      "balance_loss_clip": 0.90331352,
      "balance_loss_mlp": 1.02858949,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 2.8117500092300634,
      "language_loss": 0.77863514,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.8010205,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 2.792860984802246
    },
    {
      "auxiliary_loss_clip": 0.01202835,
      "auxiliary_loss_mlp": 0.01127521,
      "balance_loss_clip": 0.97947526,
      "balance_loss_mlp": 0.0,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 1.8712227158997772,
      "language_loss": 0.72440159,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.74770522,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 2.8233699798583984
    },
    {
      "auxiliary_loss_clip": 0.01211916,
      "auxiliary_loss_mlp": 0.01031169,
      "balance_loss_clip": 0.9857108,
      "balance_loss_mlp": 1.02193081,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 2.3544175638740925,
      "language_loss": 0.79448432,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.81691515,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.6765520572662354
    },
    {
      "auxiliary_loss_clip": 0.01219683,
      "auxiliary_loss_mlp": 0.0103278,
      "balance_loss_clip": 1.06380033,
      "balance_loss_mlp": 1.02259982,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 2.0086634945268513,
      "language_loss": 0.85290617,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.87543082,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.600090742111206
    },
    {
      "auxiliary_loss_clip": 0.01214087,
      "auxiliary_loss_mlp": 0.01034859,
      "balance_loss_clip": 0.94228625,
      "balance_loss_mlp": 1.02516675,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 4.643442684322135,
      "language_loss": 0.77411139,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.79660082,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.7566096782684326
    },
    {
      "auxiliary_loss_clip": 0.01218274,
      "auxiliary_loss_mlp": 0.01035103,
      "balance_loss_clip": 1.02175999,
      "balance_loss_mlp": 1.02541709,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 1.855378997165975,
      "language_loss": 0.71258891,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73512262,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 2.631056070327759
    },
    {
      "auxiliary_loss_clip": 0.01217871,
      "auxiliary_loss_mlp": 0.01033227,
      "balance_loss_clip": 1.02530456,
      "balance_loss_mlp": 1.02332056,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 1.7361791050543465,
      "language_loss": 0.88275105,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90526199,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.689627170562744
    },
    {
      "auxiliary_loss_clip": 0.01219532,
      "auxiliary_loss_mlp": 0.01028612,
      "balance_loss_clip": 1.02292275,
      "balance_loss_mlp": 1.01812184,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 2.684338937220168,
      "language_loss": 0.59672153,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.61920297,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 2.6710426807403564
    },
    {
      "auxiliary_loss_clip": 0.01215656,
      "auxiliary_loss_mlp": 0.01043875,
      "balance_loss_clip": 1.02311456,
      "balance_loss_mlp": 1.03344452,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.437216230986112,
      "language_loss": 0.77754164,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.80013704,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.6788017749786377
    },
    {
      "auxiliary_loss_clip": 0.01220412,
      "auxiliary_loss_mlp": 0.01037547,
      "balance_loss_clip": 1.06382048,
      "balance_loss_mlp": 1.02745581,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 3.5360146572609614,
      "language_loss": 0.75700438,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77958405,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 3.6155343055725098
    },
    {
      "auxiliary_loss_clip": 0.01215719,
      "auxiliary_loss_mlp": 0.01032671,
      "balance_loss_clip": 0.98447847,
      "balance_loss_mlp": 1.02226996,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.506065536552,
      "language_loss": 0.7942425,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81672645,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 2.726616621017456
    },
    {
      "auxiliary_loss_clip": 0.01209788,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 0.9802599,
      "balance_loss_mlp": 1.02201831,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 2.6150202429330225,
      "language_loss": 0.87398827,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89640903,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 3.7211172580718994
    },
    {
      "auxiliary_loss_clip": 0.01222052,
      "auxiliary_loss_mlp": 0.01043157,
      "balance_loss_clip": 0.9875716,
      "balance_loss_mlp": 1.03218377,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.580875157370285,
      "language_loss": 0.75646049,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.77911258,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.916771650314331
    },
    {
      "auxiliary_loss_clip": 0.01212709,
      "auxiliary_loss_mlp": 0.0104289,
      "balance_loss_clip": 0.98203754,
      "balance_loss_mlp": 1.03263164,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 1.8932952114300161,
      "language_loss": 0.86682427,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88938022,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 3.708341121673584
    },
    {
      "auxiliary_loss_clip": 0.01216393,
      "auxiliary_loss_mlp": 0.01032581,
      "balance_loss_clip": 1.02369428,
      "balance_loss_mlp": 1.02157784,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 2.283839580319867,
      "language_loss": 0.83095157,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.85344136,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.6902270317077637
    },
    {
      "auxiliary_loss_clip": 0.0121321,
      "auxiliary_loss_mlp": 0.01033629,
      "balance_loss_clip": 0.94885749,
      "balance_loss_mlp": 1.02368701,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 1.7959316484216374,
      "language_loss": 0.90299731,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.9254657,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 3.5905277729034424
    },
    {
      "auxiliary_loss_clip": 0.01202757,
      "auxiliary_loss_mlp": 0.01039222,
      "balance_loss_clip": 0.98337841,
      "balance_loss_mlp": 1.02883899,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.0836258677796122,
      "language_loss": 0.83192849,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.8543483,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.6656267642974854
    },
    {
      "auxiliary_loss_clip": 0.01215283,
      "auxiliary_loss_mlp": 0.01036885,
      "balance_loss_clip": 1.02175331,
      "balance_loss_mlp": 1.0263828,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 1.8304972457346251,
      "language_loss": 0.76477015,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.78729188,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.76265811920166
    },
    {
      "auxiliary_loss_clip": 0.01216917,
      "auxiliary_loss_mlp": 0.01043758,
      "balance_loss_clip": 0.98664343,
      "balance_loss_mlp": 1.03384018,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 1.7971765752449493,
      "language_loss": 0.83772981,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.86033654,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.645681619644165
    },
    {
      "auxiliary_loss_clip": 0.01195672,
      "auxiliary_loss_mlp": 0.01036914,
      "balance_loss_clip": 0.90197575,
      "balance_loss_mlp": 1.02651286,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 2.1378982338519,
      "language_loss": 0.86988252,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.89220834,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.762312412261963
    },
    {
      "auxiliary_loss_clip": 0.01223584,
      "auxiliary_loss_mlp": 0.01032961,
      "balance_loss_clip": 0.94595569,
      "balance_loss_mlp": 1.02296567,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 2.076357527963944,
      "language_loss": 0.67137867,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.6939441,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.7071821689605713
    },
    {
      "auxiliary_loss_clip": 0.0121605,
      "auxiliary_loss_mlp": 0.01038611,
      "balance_loss_clip": 0.94706374,
      "balance_loss_mlp": 1.02877057,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 2.241674494770584,
      "language_loss": 0.74211627,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.76466286,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.7310993671417236
    },
    {
      "auxiliary_loss_clip": 0.01219069,
      "auxiliary_loss_mlp": 0.01034131,
      "balance_loss_clip": 0.98447931,
      "balance_loss_mlp": 1.02415895,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 1.7697825136269534,
      "language_loss": 0.80276322,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.82529521,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 2.7204065322875977
    },
    {
      "auxiliary_loss_clip": 0.01217041,
      "auxiliary_loss_mlp": 0.01034801,
      "balance_loss_clip": 1.02159679,
      "balance_loss_mlp": 1.02488303,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 1.8766686171102689,
      "language_loss": 0.81801414,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.84053254,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.792598009109497
    },
    {
      "auxiliary_loss_clip": 0.01211028,
      "auxiliary_loss_mlp": 0.01040087,
      "balance_loss_clip": 0.98497558,
      "balance_loss_mlp": 1.02928114,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 1.9915692572713246,
      "language_loss": 0.83856875,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86107993,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.857550621032715
    },
    {
      "auxiliary_loss_clip": 0.01217052,
      "auxiliary_loss_mlp": 0.01034782,
      "balance_loss_clip": 0.9448489,
      "balance_loss_mlp": 1.02425003,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.616015568077329,
      "language_loss": 0.81923985,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.84175819,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.8295865058898926
    },
    {
      "auxiliary_loss_clip": 0.01205012,
      "auxiliary_loss_mlp": 0.01033282,
      "balance_loss_clip": 0.98114806,
      "balance_loss_mlp": 1.02260661,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 2.12770788048909,
      "language_loss": 0.78659773,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.80898064,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 2.703627824783325
    },
    {
      "auxiliary_loss_clip": 0.01220916,
      "auxiliary_loss_mlp": 0.01032425,
      "balance_loss_clip": 1.06383514,
      "balance_loss_mlp": 1.02205384,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 1.7876633419954093,
      "language_loss": 0.80022156,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82275498,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.554079532623291
    },
    {
      "auxiliary_loss_clip": 0.01194129,
      "auxiliary_loss_mlp": 0.01033358,
      "balance_loss_clip": 0.94388276,
      "balance_loss_mlp": 1.02248597,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 1.7312606997796445,
      "language_loss": 0.85772324,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.87999803,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 2.6896159648895264
    },
    {
      "auxiliary_loss_clip": 0.01214462,
      "auxiliary_loss_mlp": 0.01039395,
      "balance_loss_clip": 0.90491831,
      "balance_loss_mlp": 1.02913094,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 1.7566830787994114,
      "language_loss": 0.81691325,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.83945179,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 2.7446889877319336
    },
    {
      "auxiliary_loss_clip": 0.01215951,
      "auxiliary_loss_mlp": 0.01044761,
      "balance_loss_clip": 0.9875682,
      "balance_loss_mlp": 1.03467035,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 2.9144388325848793,
      "language_loss": 0.74446511,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.7670722,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.767942190170288
    },
    {
      "auxiliary_loss_clip": 0.01220877,
      "auxiliary_loss_mlp": 0.01035748,
      "balance_loss_clip": 1.06330729,
      "balance_loss_mlp": 1.02492356,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 2.740887472164384,
      "language_loss": 0.8798281,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.90239429,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 2.6476240158081055
    },
    {
      "auxiliary_loss_clip": 0.01126732,
      "auxiliary_loss_mlp": 0.01003642,
      "balance_loss_clip": 0.92315161,
      "balance_loss_mlp": 0.99944633,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8575056288435428,
      "language_loss": 0.57883912,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.60014284,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.325950860977173
    },
    {
      "auxiliary_loss_clip": 0.01223939,
      "auxiliary_loss_mlp": 0.01039524,
      "balance_loss_clip": 1.06476557,
      "balance_loss_mlp": 1.02896178,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 2.0614151082483287,
      "language_loss": 0.75566101,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.77829564,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 2.583489179611206
    },
    {
      "auxiliary_loss_clip": 0.01215928,
      "auxiliary_loss_mlp": 0.01033474,
      "balance_loss_clip": 0.94800276,
      "balance_loss_mlp": 1.02336478,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 2.3349925565186824,
      "language_loss": 0.88716733,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.90966129,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 2.6703298091888428
    },
    {
      "auxiliary_loss_clip": 0.01203402,
      "auxiliary_loss_mlp": 0.01035879,
      "balance_loss_clip": 0.9414742,
      "balance_loss_mlp": 1.02512622,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 2.2696062419383636,
      "language_loss": 0.77017003,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.79256278,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 3.685487747192383
    },
    {
      "auxiliary_loss_clip": 0.01209064,
      "auxiliary_loss_mlp": 0.01043429,
      "balance_loss_clip": 0.9871707,
      "balance_loss_mlp": 1.03187776,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.920757860743488,
      "language_loss": 0.82308829,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84561324,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 2.715961217880249
    },
    {
      "auxiliary_loss_clip": 0.0112224,
      "auxiliary_loss_mlp": 0.01004225,
      "balance_loss_clip": 0.99651366,
      "balance_loss_mlp": 0.99988598,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7841941559635668,
      "language_loss": 0.59707737,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.61834204,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 4.138270378112793
    },
    {
      "auxiliary_loss_clip": 0.01208965,
      "auxiliary_loss_mlp": 0.01045429,
      "balance_loss_clip": 0.90775371,
      "balance_loss_mlp": 1.03504014,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 1.7359689869178898,
      "language_loss": 0.92991984,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.95246375,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.793802261352539
    },
    {
      "auxiliary_loss_clip": 0.01218838,
      "auxiliary_loss_mlp": 0.01038201,
      "balance_loss_clip": 1.02640319,
      "balance_loss_mlp": 1.02717996,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 1.8365952653509392,
      "language_loss": 0.89084393,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.91341436,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 3.6719398498535156
    },
    {
      "auxiliary_loss_clip": 0.01209979,
      "auxiliary_loss_mlp": 0.01034886,
      "balance_loss_clip": 0.94385624,
      "balance_loss_mlp": 1.02404428,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 2.3578449850996814,
      "language_loss": 0.71683061,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73927927,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.7378592491149902
    },
    {
      "auxiliary_loss_clip": 0.01210249,
      "auxiliary_loss_mlp": 0.01040901,
      "balance_loss_clip": 0.94257319,
      "balance_loss_mlp": 1.02962422,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 1.8695074151172653,
      "language_loss": 0.78600031,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.80851185,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 3.6265389919281006
    },
    {
      "auxiliary_loss_clip": 0.01214828,
      "auxiliary_loss_mlp": 0.01039563,
      "balance_loss_clip": 1.02212262,
      "balance_loss_mlp": 1.02876914,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.7152736133656348,
      "language_loss": 0.8331641,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.855708,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.5819902420043945
    },
    {
      "auxiliary_loss_clip": 0.01215798,
      "auxiliary_loss_mlp": 0.01040303,
      "balance_loss_clip": 0.98839819,
      "balance_loss_mlp": 1.029199,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 1.867230006567535,
      "language_loss": 0.78039032,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.80295134,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.726759433746338
    },
    {
      "auxiliary_loss_clip": 0.01216782,
      "auxiliary_loss_mlp": 0.01037157,
      "balance_loss_clip": 1.02138937,
      "balance_loss_mlp": 1.02629685,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 1.9393776182648665,
      "language_loss": 0.83456373,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85710311,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.670869827270508
    },
    {
      "auxiliary_loss_clip": 0.01108749,
      "auxiliary_loss_mlp": 0.01003646,
      "balance_loss_clip": 0.91597414,
      "balance_loss_mlp": 0.99890196,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8198388572399478,
      "language_loss": 0.6000222,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.62114614,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.252830743789673
    },
    {
      "auxiliary_loss_clip": 0.01215862,
      "auxiliary_loss_mlp": 0.01127528,
      "balance_loss_clip": 1.02207398,
      "balance_loss_mlp": 0.0,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 2.3953391188216404,
      "language_loss": 0.84292364,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.86635756,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.664425849914551
    },
    {
      "auxiliary_loss_clip": 0.01214684,
      "auxiliary_loss_mlp": 0.01127888,
      "balance_loss_clip": 0.94610995,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.6183482129088755,
      "language_loss": 0.86178291,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.88520867,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.7731430530548096
    },
    {
      "auxiliary_loss_clip": 0.01221616,
      "auxiliary_loss_mlp": 0.01042272,
      "balance_loss_clip": 1.02385759,
      "balance_loss_mlp": 1.0318768,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 1.9402109928189102,
      "language_loss": 0.8035779,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82621682,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.6803019046783447
    },
    {
      "auxiliary_loss_clip": 0.01215776,
      "auxiliary_loss_mlp": 0.01045,
      "balance_loss_clip": 1.02361679,
      "balance_loss_mlp": 1.03427148,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 1.9837090592368354,
      "language_loss": 0.86774218,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.89034998,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.6157443523406982
    },
    {
      "auxiliary_loss_clip": 0.01217486,
      "auxiliary_loss_mlp": 0.01038874,
      "balance_loss_clip": 1.02397907,
      "balance_loss_mlp": 1.02830005,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 2.014366255666369,
      "language_loss": 0.85761118,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.88017476,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.6619598865509033
    },
    {
      "auxiliary_loss_clip": 0.01210833,
      "auxiliary_loss_mlp": 0.01033168,
      "balance_loss_clip": 0.94450504,
      "balance_loss_mlp": 1.0214678,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 2.20417440627985,
      "language_loss": 0.7606976,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78313762,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.8959524631500244
    },
    {
      "auxiliary_loss_clip": 0.01114214,
      "auxiliary_loss_mlp": 0.01008182,
      "balance_loss_clip": 0.99537158,
      "balance_loss_mlp": 1.00348496,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8645638114033127,
      "language_loss": 0.61892998,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.64015394,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 3.3428053855895996
    },
    {
      "auxiliary_loss_clip": 0.0121996,
      "auxiliary_loss_mlp": 0.0112796,
      "balance_loss_clip": 1.01959276,
      "balance_loss_mlp": 0.0,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 1.6753318693751087,
      "language_loss": 0.84099972,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.86447895,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 2.7322332859039307
    },
    {
      "auxiliary_loss_clip": 0.01212135,
      "auxiliary_loss_mlp": 0.01042561,
      "balance_loss_clip": 1.02271199,
      "balance_loss_mlp": 1.03181458,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 1.8245824850705508,
      "language_loss": 0.85945368,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88200068,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.659755229949951
    },
    {
      "auxiliary_loss_clip": 0.01214667,
      "auxiliary_loss_mlp": 0.01044362,
      "balance_loss_clip": 1.02252054,
      "balance_loss_mlp": 1.03396702,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 5.716883094099848,
      "language_loss": 0.78324616,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80583644,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 2.690824508666992
    },
    {
      "auxiliary_loss_clip": 0.01218381,
      "auxiliary_loss_mlp": 0.01038671,
      "balance_loss_clip": 1.02201414,
      "balance_loss_mlp": 1.02732253,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 1.8349322736684182,
      "language_loss": 0.84871757,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.87128806,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 2.668069839477539
    },
    {
      "auxiliary_loss_clip": 0.01114123,
      "auxiliary_loss_mlp": 0.01002356,
      "balance_loss_clip": 0.87480783,
      "balance_loss_mlp": 0.99761105,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7897502394467226,
      "language_loss": 0.55971444,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.58087921,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 3.1620352268218994
    },
    {
      "auxiliary_loss_clip": 0.01206677,
      "auxiliary_loss_mlp": 0.01034283,
      "balance_loss_clip": 0.90052986,
      "balance_loss_mlp": 1.02332759,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 1.9501898930352235,
      "language_loss": 0.87422323,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.89663285,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.7941298484802246
    },
    {
      "auxiliary_loss_clip": 0.01215748,
      "auxiliary_loss_mlp": 0.01040564,
      "balance_loss_clip": 0.98665047,
      "balance_loss_mlp": 1.0300442,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.006444244233659,
      "language_loss": 0.75136018,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.77392328,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 2.7061219215393066
    },
    {
      "auxiliary_loss_clip": 0.01208785,
      "auxiliary_loss_mlp": 0.01037528,
      "balance_loss_clip": 0.94811642,
      "balance_loss_mlp": 1.02670991,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 1.613065271549423,
      "language_loss": 0.8313309,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.85379404,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 2.6821341514587402
    },
    {
      "auxiliary_loss_clip": 0.01211966,
      "auxiliary_loss_mlp": 0.01038694,
      "balance_loss_clip": 0.98195159,
      "balance_loss_mlp": 1.02763772,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 1.5931348297361272,
      "language_loss": 0.70947999,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.73198664,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 3.6574411392211914
    },
    {
      "auxiliary_loss_clip": 0.01220871,
      "auxiliary_loss_mlp": 0.01038825,
      "balance_loss_clip": 1.02284098,
      "balance_loss_mlp": 1.02835286,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 3.7071957677873932,
      "language_loss": 0.81143105,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83402801,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 2.5596776008605957
    },
    {
      "auxiliary_loss_clip": 0.01212035,
      "auxiliary_loss_mlp": 0.01039094,
      "balance_loss_clip": 1.02282631,
      "balance_loss_mlp": 1.02924156,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.4465547960145804,
      "language_loss": 0.74602175,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76853299,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 3.4823360443115234
    },
    {
      "auxiliary_loss_clip": 0.01221345,
      "auxiliary_loss_mlp": 0.01127788,
      "balance_loss_clip": 0.94429624,
      "balance_loss_mlp": 0.0,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 2.180111665533421,
      "language_loss": 0.79047871,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.81397009,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 3.584535837173462
    },
    {
      "auxiliary_loss_clip": 0.01206338,
      "auxiliary_loss_mlp": 0.010419,
      "balance_loss_clip": 0.98033381,
      "balance_loss_mlp": 1.03098631,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 4.983087098107026,
      "language_loss": 0.81446946,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83695185,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.672992467880249
    },
    {
      "auxiliary_loss_clip": 0.01218615,
      "auxiliary_loss_mlp": 0.01038026,
      "balance_loss_clip": 1.06135416,
      "balance_loss_mlp": 1.02695751,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.674919746154901,
      "language_loss": 0.84715176,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.86971825,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 3.5059814453125
    },
    {
      "auxiliary_loss_clip": 0.01204806,
      "auxiliary_loss_mlp": 0.01050748,
      "balance_loss_clip": 0.98420906,
      "balance_loss_mlp": 1.04025769,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 2.591009708153044,
      "language_loss": 0.81432605,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.83688164,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.679051399230957
    },
    {
      "auxiliary_loss_clip": 0.01202,
      "auxiliary_loss_mlp": 0.01030718,
      "balance_loss_clip": 0.94038451,
      "balance_loss_mlp": 1.02093053,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 2.5825672287215418,
      "language_loss": 0.78259039,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80491757,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.644334554672241
    },
    {
      "auxiliary_loss_clip": 0.01215775,
      "auxiliary_loss_mlp": 0.01042407,
      "balance_loss_clip": 1.06252038,
      "balance_loss_mlp": 1.03150511,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 1.7522583562120535,
      "language_loss": 0.80677491,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.82935673,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.5741357803344727
    },
    {
      "auxiliary_loss_clip": 0.0121112,
      "auxiliary_loss_mlp": 0.01037994,
      "balance_loss_clip": 0.97982466,
      "balance_loss_mlp": 1.02693689,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.6179592151941975,
      "language_loss": 0.82445461,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.84694576,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.6497552394866943
    },
    {
      "auxiliary_loss_clip": 0.01207307,
      "auxiliary_loss_mlp": 0.01036953,
      "balance_loss_clip": 0.98216021,
      "balance_loss_mlp": 1.02538383,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 2.4148338355289427,
      "language_loss": 0.83470529,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.85714793,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.6372573375701904
    },
    {
      "auxiliary_loss_clip": 0.01210363,
      "auxiliary_loss_mlp": 0.01039088,
      "balance_loss_clip": 0.94100314,
      "balance_loss_mlp": 1.02834094,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 1.7611941574730323,
      "language_loss": 0.78441983,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.80691439,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.6950745582580566
    },
    {
      "auxiliary_loss_clip": 0.01212467,
      "auxiliary_loss_mlp": 0.01035417,
      "balance_loss_clip": 1.02353001,
      "balance_loss_mlp": 1.02474821,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.599358833587349,
      "language_loss": 0.77452672,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.79700553,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.6922199726104736
    },
    {
      "auxiliary_loss_clip": 0.01213307,
      "auxiliary_loss_mlp": 0.01037036,
      "balance_loss_clip": 0.98173881,
      "balance_loss_mlp": 1.02644467,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 3.10302986803653,
      "language_loss": 0.74692345,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.76942688,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.581761360168457
    },
    {
      "auxiliary_loss_clip": 0.01213009,
      "auxiliary_loss_mlp": 0.01042076,
      "balance_loss_clip": 0.98605847,
      "balance_loss_mlp": 1.03168702,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 1.8368940205450086,
      "language_loss": 0.81753242,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.8400833,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.6355533599853516
    },
    {
      "auxiliary_loss_clip": 0.01200517,
      "auxiliary_loss_mlp": 0.01034609,
      "balance_loss_clip": 0.98282987,
      "balance_loss_mlp": 1.02361202,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 1.9609975214893327,
      "language_loss": 0.74790013,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.77025133,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.6948978900909424
    },
    {
      "auxiliary_loss_clip": 0.01214483,
      "auxiliary_loss_mlp": 0.01040449,
      "balance_loss_clip": 0.98159605,
      "balance_loss_mlp": 1.02989888,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 1.6060523499498933,
      "language_loss": 0.72778201,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.75033134,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.776090621948242
    },
    {
      "auxiliary_loss_clip": 0.01218867,
      "auxiliary_loss_mlp": 0.01032716,
      "balance_loss_clip": 1.06010938,
      "balance_loss_mlp": 1.02212453,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 2.025058212413782,
      "language_loss": 0.77429909,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.79681492,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.596721887588501
    },
    {
      "auxiliary_loss_clip": 0.01206037,
      "auxiliary_loss_mlp": 0.01035294,
      "balance_loss_clip": 1.01989412,
      "balance_loss_mlp": 1.02504253,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 1.9771595742940276,
      "language_loss": 0.79421771,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.81663108,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 2.7684051990509033
    },
    {
      "auxiliary_loss_clip": 0.0121257,
      "auxiliary_loss_mlp": 0.01037963,
      "balance_loss_clip": 0.98770714,
      "balance_loss_mlp": 1.02805114,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 1.850970348721938,
      "language_loss": 0.80491155,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.8274169,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 2.8265130519866943
    },
    {
      "auxiliary_loss_clip": 0.01110425,
      "auxiliary_loss_mlp": 0.01014361,
      "balance_loss_clip": 0.87321192,
      "balance_loss_mlp": 1.00995016,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.9897887201401904,
      "language_loss": 0.63170695,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65295482,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 3.2333977222442627
    },
    {
      "auxiliary_loss_clip": 0.01208284,
      "auxiliary_loss_mlp": 0.0104108,
      "balance_loss_clip": 0.94535172,
      "balance_loss_mlp": 1.02970755,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 1.7692185627993409,
      "language_loss": 0.71810102,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74059469,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 2.7263059616088867
    },
    {
      "auxiliary_loss_clip": 0.01217449,
      "auxiliary_loss_mlp": 0.01035898,
      "balance_loss_clip": 0.98527884,
      "balance_loss_mlp": 1.02637947,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 2.004563455283112,
      "language_loss": 0.8199116,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.84244502,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.645073890686035
    },
    {
      "auxiliary_loss_clip": 0.01219491,
      "auxiliary_loss_mlp": 0.01037412,
      "balance_loss_clip": 1.06270444,
      "balance_loss_mlp": 1.02624798,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 2.403614806104032,
      "language_loss": 0.87382042,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.89638942,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.579712390899658
    },
    {
      "auxiliary_loss_clip": 0.01217188,
      "auxiliary_loss_mlp": 0.01031389,
      "balance_loss_clip": 1.02064395,
      "balance_loss_mlp": 1.02152407,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 1.9319551735243758,
      "language_loss": 0.84892267,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.8714084,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 2.62536358833313
    },
    {
      "auxiliary_loss_clip": 0.01210296,
      "auxiliary_loss_mlp": 0.01039204,
      "balance_loss_clip": 0.90914059,
      "balance_loss_mlp": 1.02755737,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.5298054254895437,
      "language_loss": 0.79875779,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.82125282,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 2.7563912868499756
    },
    {
      "auxiliary_loss_clip": 0.012197,
      "auxiliary_loss_mlp": 0.01038725,
      "balance_loss_clip": 1.02627766,
      "balance_loss_mlp": 1.02883101,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 1.5785901465818328,
      "language_loss": 0.81208766,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83467191,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 3.6082568168640137
    },
    {
      "auxiliary_loss_clip": 0.01204539,
      "auxiliary_loss_mlp": 0.01039316,
      "balance_loss_clip": 0.91175848,
      "balance_loss_mlp": 1.02886093,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 2.29240647678699,
      "language_loss": 0.80743724,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.82987583,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 2.7337892055511475
    },
    {
      "auxiliary_loss_clip": 0.01208479,
      "auxiliary_loss_mlp": 0.01034756,
      "balance_loss_clip": 1.02065063,
      "balance_loss_mlp": 1.02496922,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 1.9456363745833425,
      "language_loss": 0.95454681,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.97697914,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 3.7466399669647217
    },
    {
      "auxiliary_loss_clip": 0.01202218,
      "auxiliary_loss_mlp": 0.01034521,
      "balance_loss_clip": 0.86112958,
      "balance_loss_mlp": 1.02434683,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 1.666990630028351,
      "language_loss": 0.78462601,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.80699342,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 3.725151300430298
    },
    {
      "auxiliary_loss_clip": 0.01212907,
      "auxiliary_loss_mlp": 0.01045176,
      "balance_loss_clip": 1.02432024,
      "balance_loss_mlp": 1.03476918,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 2.0664132858801727,
      "language_loss": 0.78559315,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.80817389,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.6283891201019287
    },
    {
      "auxiliary_loss_clip": 0.01209405,
      "auxiliary_loss_mlp": 0.01040728,
      "balance_loss_clip": 0.98356628,
      "balance_loss_mlp": 1.03048754,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 2.2406909099919567,
      "language_loss": 0.77656394,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.79906535,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.76967453956604
    },
    {
      "auxiliary_loss_clip": 0.01220122,
      "auxiliary_loss_mlp": 0.01033045,
      "balance_loss_clip": 1.02392411,
      "balance_loss_mlp": 1.02310252,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.3093858654493937,
      "language_loss": 0.73201311,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.75454473,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 3.5535247325897217
    },
    {
      "auxiliary_loss_clip": 0.01201038,
      "auxiliary_loss_mlp": 0.0112764,
      "balance_loss_clip": 0.98125499,
      "balance_loss_mlp": 0.0,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 1.9475104203341809,
      "language_loss": 0.7955637,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.81885046,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.7309932708740234
    },
    {
      "auxiliary_loss_clip": 0.01210706,
      "auxiliary_loss_mlp": 0.01127462,
      "balance_loss_clip": 0.98557258,
      "balance_loss_mlp": 0.0,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 2.3194530394864548,
      "language_loss": 0.77736521,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.80074686,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.6737451553344727
    },
    {
      "auxiliary_loss_clip": 0.01101061,
      "auxiliary_loss_mlp": 0.0100639,
      "balance_loss_clip": 0.94738829,
      "balance_loss_mlp": 1.00212193,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7763851295987985,
      "language_loss": 0.59451348,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61558795,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.2345902919769287
    },
    {
      "auxiliary_loss_clip": 0.01217226,
      "auxiliary_loss_mlp": 0.0103752,
      "balance_loss_clip": 1.0610801,
      "balance_loss_mlp": 1.02741694,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 1.9281862496026132,
      "language_loss": 0.82737035,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.84991789,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.6380598545074463
    },
    {
      "auxiliary_loss_clip": 0.01198476,
      "auxiliary_loss_mlp": 0.01036414,
      "balance_loss_clip": 0.90559697,
      "balance_loss_mlp": 1.02615643,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.8549241907633132,
      "language_loss": 0.8342635,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.85661244,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.7462158203125
    },
    {
      "auxiliary_loss_clip": 0.01217386,
      "auxiliary_loss_mlp": 0.01038123,
      "balance_loss_clip": 1.06217384,
      "balance_loss_mlp": 1.02763867,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 2.059400334836705,
      "language_loss": 0.89818323,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92073834,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.664259195327759
    },
    {
      "auxiliary_loss_clip": 0.0121728,
      "auxiliary_loss_mlp": 0.01033671,
      "balance_loss_clip": 1.02145481,
      "balance_loss_mlp": 1.02340102,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 2.034291009408028,
      "language_loss": 0.80755448,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.83006406,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.663846969604492
    },
    {
      "auxiliary_loss_clip": 0.01213817,
      "auxiliary_loss_mlp": 0.01039265,
      "balance_loss_clip": 1.02216673,
      "balance_loss_mlp": 1.02912068,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 1.6276119768515436,
      "language_loss": 0.73247528,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75500607,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.664144515991211
    },
    {
      "auxiliary_loss_clip": 0.01210977,
      "auxiliary_loss_mlp": 0.01039067,
      "balance_loss_clip": 1.02109861,
      "balance_loss_mlp": 1.02913082,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 2.197761418158773,
      "language_loss": 0.75503671,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.77753711,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.599184274673462
    },
    {
      "auxiliary_loss_clip": 0.01219263,
      "auxiliary_loss_mlp": 0.010427,
      "balance_loss_clip": 1.06135607,
      "balance_loss_mlp": 1.03197098,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 2.2823022552300354,
      "language_loss": 0.82349694,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84611654,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.6023504734039307
    },
    {
      "auxiliary_loss_clip": 0.01215761,
      "auxiliary_loss_mlp": 0.01042002,
      "balance_loss_clip": 1.02356505,
      "balance_loss_mlp": 1.03143954,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 1.952364315215956,
      "language_loss": 0.79937774,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82195532,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.67764949798584
    },
    {
      "auxiliary_loss_clip": 0.01212974,
      "auxiliary_loss_mlp": 0.01033611,
      "balance_loss_clip": 1.02273989,
      "balance_loss_mlp": 1.0240798,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 2.2508993069988823,
      "language_loss": 0.84179997,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86426586,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 2.6375882625579834
    },
    {
      "auxiliary_loss_clip": 0.0119867,
      "auxiliary_loss_mlp": 0.01043792,
      "balance_loss_clip": 0.86685437,
      "balance_loss_mlp": 1.03259873,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 1.9482667420849256,
      "language_loss": 0.82896769,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.85139233,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.7066245079040527
    },
    {
      "auxiliary_loss_clip": 0.01211332,
      "auxiliary_loss_mlp": 0.0103897,
      "balance_loss_clip": 0.94761533,
      "balance_loss_mlp": 1.02864075,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 1.8837732589132605,
      "language_loss": 0.75169766,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.77420068,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 2.7538487911224365
    },
    {
      "auxiliary_loss_clip": 0.01217279,
      "auxiliary_loss_mlp": 0.01047653,
      "balance_loss_clip": 1.02311635,
      "balance_loss_mlp": 1.03758013,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 1.945948600198489,
      "language_loss": 0.78088582,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.8035351,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 2.7039451599121094
    },
    {
      "auxiliary_loss_clip": 0.01208956,
      "auxiliary_loss_mlp": 0.01039677,
      "balance_loss_clip": 0.94493586,
      "balance_loss_mlp": 1.02935362,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.7217540630995485,
      "language_loss": 0.80666232,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.82914865,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.8332149982452393
    },
    {
      "auxiliary_loss_clip": 0.01205627,
      "auxiliary_loss_mlp": 0.01036448,
      "balance_loss_clip": 1.01927114,
      "balance_loss_mlp": 1.02572513,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 1.8493234616597611,
      "language_loss": 0.76860321,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.79102397,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.656853199005127
    },
    {
      "auxiliary_loss_clip": 0.01208899,
      "auxiliary_loss_mlp": 0.01036695,
      "balance_loss_clip": 0.93949604,
      "balance_loss_mlp": 1.02666962,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 1.6890867257273277,
      "language_loss": 0.81296247,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83541846,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 2.767711877822876
    },
    {
      "auxiliary_loss_clip": 0.01218715,
      "auxiliary_loss_mlp": 0.01041569,
      "balance_loss_clip": 1.06401706,
      "balance_loss_mlp": 1.03110838,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 1.8189641496204378,
      "language_loss": 0.76240432,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78500712,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 3.5578157901763916
    },
    {
      "auxiliary_loss_clip": 0.01212128,
      "auxiliary_loss_mlp": 0.01042004,
      "balance_loss_clip": 0.98326302,
      "balance_loss_mlp": 1.0318538,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.831232890428992,
      "language_loss": 0.83589303,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85843444,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 2.7721405029296875
    },
    {
      "auxiliary_loss_clip": 0.01179491,
      "auxiliary_loss_mlp": 0.01041989,
      "balance_loss_clip": 0.82115376,
      "balance_loss_mlp": 1.03103924,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 1.8172465403117475,
      "language_loss": 0.80864716,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83086193,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 2.8575026988983154
    },
    {
      "auxiliary_loss_clip": 0.01216499,
      "auxiliary_loss_mlp": 0.01037543,
      "balance_loss_clip": 1.06193662,
      "balance_loss_mlp": 1.0272727,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.5124383637260697,
      "language_loss": 0.85163391,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.8741743,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 3.605942726135254
    },
    {
      "auxiliary_loss_clip": 0.01209219,
      "auxiliary_loss_mlp": 0.01043648,
      "balance_loss_clip": 1.02193022,
      "balance_loss_mlp": 1.03311634,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.7683885734943354,
      "language_loss": 0.8032465,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.82577515,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 3.7141013145446777
    },
    {
      "auxiliary_loss_clip": 0.01213495,
      "auxiliary_loss_mlp": 0.01041631,
      "balance_loss_clip": 0.98493689,
      "balance_loss_mlp": 1.03168344,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 1.7599399503171453,
      "language_loss": 0.77491319,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79746443,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.76835560798645
    },
    {
      "auxiliary_loss_clip": 0.01199839,
      "auxiliary_loss_mlp": 0.01035629,
      "balance_loss_clip": 0.90652144,
      "balance_loss_mlp": 1.02551425,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 1.763640338042746,
      "language_loss": 0.76257956,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78493428,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 2.7615461349487305
    },
    {
      "auxiliary_loss_clip": 0.01217172,
      "auxiliary_loss_mlp": 0.01128103,
      "balance_loss_clip": 1.06119382,
      "balance_loss_mlp": 0.0,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.0691856155511954,
      "language_loss": 0.81886888,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.84232163,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 3.5144081115722656
    },
    {
      "auxiliary_loss_clip": 0.012115,
      "auxiliary_loss_mlp": 0.01037039,
      "balance_loss_clip": 0.90406448,
      "balance_loss_mlp": 1.02694225,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 1.8336112311166088,
      "language_loss": 0.783764,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80624938,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.8388330936431885
    },
    {
      "auxiliary_loss_clip": 0.01212298,
      "auxiliary_loss_mlp": 0.01030304,
      "balance_loss_clip": 1.059484,
      "balance_loss_mlp": 1.02125573,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 1.9763987159070528,
      "language_loss": 0.85026228,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87268835,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.7265496253967285
    },
    {
      "auxiliary_loss_clip": 0.01109299,
      "auxiliary_loss_mlp": 0.01013196,
      "balance_loss_clip": 0.91263384,
      "balance_loss_mlp": 1.00885713,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.6998809269503627,
      "language_loss": 0.59202152,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.6132465,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.30880069732666
    },
    {
      "auxiliary_loss_clip": 0.0121634,
      "auxiliary_loss_mlp": 0.01033741,
      "balance_loss_clip": 0.98395532,
      "balance_loss_mlp": 1.02353048,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 2.2585590068907577,
      "language_loss": 0.64595646,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.66845727,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.7554516792297363
    },
    {
      "auxiliary_loss_clip": 0.01209148,
      "auxiliary_loss_mlp": 0.01027789,
      "balance_loss_clip": 0.945907,
      "balance_loss_mlp": 1.01748335,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 1.7319219657268796,
      "language_loss": 0.6496588,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67202818,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.7513651847839355
    },
    {
      "auxiliary_loss_clip": 0.0120682,
      "auxiliary_loss_mlp": 0.01030498,
      "balance_loss_clip": 0.98448813,
      "balance_loss_mlp": 1.02115178,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 1.8578371014823223,
      "language_loss": 0.89549696,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91787016,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.7715394496917725
    },
    {
      "auxiliary_loss_clip": 0.01198365,
      "auxiliary_loss_mlp": 0.01034482,
      "balance_loss_clip": 0.89989126,
      "balance_loss_mlp": 1.02451646,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 2.1156260495320334,
      "language_loss": 0.90252304,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92485148,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.806830883026123
    },
    {
      "auxiliary_loss_clip": 0.01215244,
      "auxiliary_loss_mlp": 0.01031492,
      "balance_loss_clip": 1.02007651,
      "balance_loss_mlp": 1.02138925,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 1.7210182726579653,
      "language_loss": 0.70477432,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.72724169,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.797985315322876
    },
    {
      "auxiliary_loss_clip": 0.01208599,
      "auxiliary_loss_mlp": 0.0103402,
      "balance_loss_clip": 1.01800346,
      "balance_loss_mlp": 1.02423918,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.08608173652339,
      "language_loss": 0.8140136,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.83643979,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.739868402481079
    },
    {
      "auxiliary_loss_clip": 0.01212209,
      "auxiliary_loss_mlp": 0.01039509,
      "balance_loss_clip": 0.98733306,
      "balance_loss_mlp": 1.02873898,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 1.5968036751042078,
      "language_loss": 0.77854759,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.80106479,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.718656063079834
    },
    {
      "auxiliary_loss_clip": 0.01207486,
      "auxiliary_loss_mlp": 0.0103183,
      "balance_loss_clip": 0.93903673,
      "balance_loss_mlp": 1.02258492,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 1.9187332842224774,
      "language_loss": 0.80512542,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.82751858,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.755401849746704
    },
    {
      "auxiliary_loss_clip": 0.01212465,
      "auxiliary_loss_mlp": 0.01036312,
      "balance_loss_clip": 1.02298152,
      "balance_loss_mlp": 1.02679312,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 1.8954800358121255,
      "language_loss": 0.81292284,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83541059,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.652951955795288
    },
    {
      "auxiliary_loss_clip": 0.01107736,
      "auxiliary_loss_mlp": 0.01004886,
      "balance_loss_clip": 0.98522222,
      "balance_loss_mlp": 1.00066555,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.8301963022591884,
      "language_loss": 0.61056173,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63168788,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.246316432952881
    },
    {
      "auxiliary_loss_clip": 0.01198009,
      "auxiliary_loss_mlp": 0.01034116,
      "balance_loss_clip": 0.98092687,
      "balance_loss_mlp": 1.02435303,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.4982272008611401,
      "language_loss": 0.75379902,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77612031,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.6650164127349854
    },
    {
      "auxiliary_loss_clip": 0.01203268,
      "auxiliary_loss_mlp": 0.01032474,
      "balance_loss_clip": 0.97916126,
      "balance_loss_mlp": 1.02198374,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 1.8447589843287906,
      "language_loss": 0.77729905,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.79965651,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 2.6877050399780273
    },
    {
      "auxiliary_loss_clip": 0.01197943,
      "auxiliary_loss_mlp": 0.01032165,
      "balance_loss_clip": 0.97790974,
      "balance_loss_mlp": 1.02118015,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 2.1297738151001457,
      "language_loss": 0.76831496,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.79061604,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 2.6728317737579346
    },
    {
      "auxiliary_loss_clip": 0.01206975,
      "auxiliary_loss_mlp": 0.01042507,
      "balance_loss_clip": 0.98125476,
      "balance_loss_mlp": 1.03221381,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 1.839314635052048,
      "language_loss": 0.82984686,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.85234171,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.713763475418091
    },
    {
      "auxiliary_loss_clip": 0.01199896,
      "auxiliary_loss_mlp": 0.01036144,
      "balance_loss_clip": 0.98032743,
      "balance_loss_mlp": 1.02622604,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 1.7287552782953661,
      "language_loss": 0.8680169,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89037734,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.6717476844787598
    },
    {
      "auxiliary_loss_clip": 0.0121691,
      "auxiliary_loss_mlp": 0.01037044,
      "balance_loss_clip": 1.06171036,
      "balance_loss_mlp": 1.02706671,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 1.7052463129298254,
      "language_loss": 0.7565757,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.77911526,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.6320934295654297
    },
    {
      "auxiliary_loss_clip": 0.0120742,
      "auxiliary_loss_mlp": 0.01127456,
      "balance_loss_clip": 0.90363884,
      "balance_loss_mlp": 0.0,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 1.8148545791268653,
      "language_loss": 0.86238891,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.88573766,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 3.71750807762146
    },
    {
      "auxiliary_loss_clip": 0.01208845,
      "auxiliary_loss_mlp": 0.0102673,
      "balance_loss_clip": 1.02027047,
      "balance_loss_mlp": 1.01675177,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 1.844803698166738,
      "language_loss": 0.82768768,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.85004348,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 2.658571481704712
    },
    {
      "auxiliary_loss_clip": 0.01197637,
      "auxiliary_loss_mlp": 0.01035541,
      "balance_loss_clip": 0.94066525,
      "balance_loss_mlp": 1.02488947,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 1.9464139185200222,
      "language_loss": 0.75993145,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.78226316,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 3.6585910320281982
    },
    {
      "auxiliary_loss_clip": 0.01207313,
      "auxiliary_loss_mlp": 0.01035658,
      "balance_loss_clip": 0.97932088,
      "balance_loss_mlp": 1.02560878,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 1.4417725673619617,
      "language_loss": 0.81488031,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83730996,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 3.8107190132141113
    },
    {
      "auxiliary_loss_clip": 0.01215146,
      "auxiliary_loss_mlp": 0.01043939,
      "balance_loss_clip": 0.90816796,
      "balance_loss_mlp": 1.03352642,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.8755259062710667,
      "language_loss": 0.76940465,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79199553,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 2.695662021636963
    },
    {
      "auxiliary_loss_clip": 0.01205723,
      "auxiliary_loss_mlp": 0.01041836,
      "balance_loss_clip": 0.98039544,
      "balance_loss_mlp": 1.03203702,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 2.045296200668218,
      "language_loss": 0.85847104,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88094664,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 3.6315832138061523
    },
    {
      "auxiliary_loss_clip": 0.01208948,
      "auxiliary_loss_mlp": 0.01033586,
      "balance_loss_clip": 0.9400413,
      "balance_loss_mlp": 1.02342939,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.39019840441103,
      "language_loss": 0.82733518,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.84976053,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.7070910930633545
    },
    {
      "auxiliary_loss_clip": 0.01211091,
      "auxiliary_loss_mlp": 0.01035724,
      "balance_loss_clip": 0.98202497,
      "balance_loss_mlp": 1.02528143,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 3.4591592179476036,
      "language_loss": 0.78545666,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.80792487,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.681774377822876
    },
    {
      "auxiliary_loss_clip": 0.01220679,
      "auxiliary_loss_mlp": 0.01035301,
      "balance_loss_clip": 1.02652347,
      "balance_loss_mlp": 1.02521563,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 1.9488201760080943,
      "language_loss": 0.8707363,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.89329606,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.675969362258911
    },
    {
      "auxiliary_loss_clip": 0.01195543,
      "auxiliary_loss_mlp": 0.01038892,
      "balance_loss_clip": 0.89840984,
      "balance_loss_mlp": 1.02919459,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 2.275734181934113,
      "language_loss": 0.71446109,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73680544,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.718625783920288
    },
    {
      "auxiliary_loss_clip": 0.01204889,
      "auxiliary_loss_mlp": 0.01038168,
      "balance_loss_clip": 1.01744282,
      "balance_loss_mlp": 1.02813697,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 2.0530997091131753,
      "language_loss": 0.80835503,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.83078563,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.751965284347534
    },
    {
      "auxiliary_loss_clip": 0.01217913,
      "auxiliary_loss_mlp": 0.01034453,
      "balance_loss_clip": 1.06092143,
      "balance_loss_mlp": 1.02354503,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 2.8484741844510006,
      "language_loss": 0.69281769,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71534145,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.6714205741882324
    },
    {
      "auxiliary_loss_clip": 0.0121371,
      "auxiliary_loss_mlp": 0.01032566,
      "balance_loss_clip": 1.06013322,
      "balance_loss_mlp": 1.02234936,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 1.8564597674023737,
      "language_loss": 0.65993214,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68239492,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.7449913024902344
    },
    {
      "auxiliary_loss_clip": 0.01213317,
      "auxiliary_loss_mlp": 0.01042787,
      "balance_loss_clip": 1.06040192,
      "balance_loss_mlp": 1.03298783,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 2.4930602328636167,
      "language_loss": 0.85893649,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.8814975,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.590020179748535
    },
    {
      "auxiliary_loss_clip": 0.01206813,
      "auxiliary_loss_mlp": 0.0103638,
      "balance_loss_clip": 1.0204165,
      "balance_loss_mlp": 1.02683127,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.074376100984681,
      "language_loss": 0.7877425,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.81017447,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.6824162006378174
    },
    {
      "auxiliary_loss_clip": 0.0120009,
      "auxiliary_loss_mlp": 0.01040794,
      "balance_loss_clip": 0.93910635,
      "balance_loss_mlp": 1.03061974,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.161253152585186,
      "language_loss": 0.74611914,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.76852798,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.693695545196533
    },
    {
      "auxiliary_loss_clip": 0.01209324,
      "auxiliary_loss_mlp": 0.01041209,
      "balance_loss_clip": 0.9804188,
      "balance_loss_mlp": 1.03125536,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 2.150118497703898,
      "language_loss": 0.87480235,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89730763,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 2.749594211578369
    },
    {
      "auxiliary_loss_clip": 0.0119565,
      "auxiliary_loss_mlp": 0.01037744,
      "balance_loss_clip": 0.90093046,
      "balance_loss_mlp": 1.02744389,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 2.151880092763126,
      "language_loss": 0.7076723,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.73000622,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.7514779567718506
    },
    {
      "auxiliary_loss_clip": 0.01210065,
      "auxiliary_loss_mlp": 0.01034585,
      "balance_loss_clip": 0.94379902,
      "balance_loss_mlp": 1.02497077,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 1.9089841664834843,
      "language_loss": 0.91856849,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.94101501,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.7703940868377686
    },
    {
      "auxiliary_loss_clip": 0.01217071,
      "auxiliary_loss_mlp": 0.01127887,
      "balance_loss_clip": 1.02264452,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 1.789435922194459,
      "language_loss": 0.80145985,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.82490945,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.7591898441314697
    },
    {
      "auxiliary_loss_clip": 0.01209485,
      "auxiliary_loss_mlp": 0.01038351,
      "balance_loss_clip": 1.01945686,
      "balance_loss_mlp": 1.02926755,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 2.5504193068903,
      "language_loss": 0.8492015,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.8716799,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 2.6441385746002197
    },
    {
      "auxiliary_loss_clip": 0.01193169,
      "auxiliary_loss_mlp": 0.01033072,
      "balance_loss_clip": 0.94047964,
      "balance_loss_mlp": 1.02374434,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.359873460227759,
      "language_loss": 0.77553368,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79779613,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 2.779238700866699
    },
    {
      "auxiliary_loss_clip": 0.01202707,
      "auxiliary_loss_mlp": 0.01037264,
      "balance_loss_clip": 0.94287819,
      "balance_loss_mlp": 1.02646339,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 2.0750082698607186,
      "language_loss": 0.80890119,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.83130091,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 2.704019784927368
    },
    {
      "auxiliary_loss_clip": 0.01195348,
      "auxiliary_loss_mlp": 0.01034813,
      "balance_loss_clip": 1.01601863,
      "balance_loss_mlp": 1.0254848,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 1.6652690124032232,
      "language_loss": 0.75002891,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.77233052,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 2.7560272216796875
    },
    {
      "auxiliary_loss_clip": 0.0121265,
      "auxiliary_loss_mlp": 0.01040735,
      "balance_loss_clip": 1.05930734,
      "balance_loss_mlp": 1.03098989,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 2.3228800945337524,
      "language_loss": 0.81598735,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.83852124,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.619534730911255
    },
    {
      "auxiliary_loss_clip": 0.01213943,
      "auxiliary_loss_mlp": 0.01127305,
      "balance_loss_clip": 0.98568559,
      "balance_loss_mlp": 0.0,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 1.591926207350638,
      "language_loss": 0.81462169,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.83803415,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 2.798651933670044
    },
    {
      "auxiliary_loss_clip": 0.0120969,
      "auxiliary_loss_mlp": 0.01040623,
      "balance_loss_clip": 1.01976657,
      "balance_loss_mlp": 1.03086603,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 1.7474489967435665,
      "language_loss": 0.83863151,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.86113465,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 3.6784868240356445
    },
    {
      "auxiliary_loss_clip": 0.01213007,
      "auxiliary_loss_mlp": 0.01032085,
      "balance_loss_clip": 1.0591687,
      "balance_loss_mlp": 1.02133822,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 2.1062998088135783,
      "language_loss": 0.85178757,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87423849,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 2.5678651332855225
    },
    {
      "auxiliary_loss_clip": 0.01216317,
      "auxiliary_loss_mlp": 0.01030783,
      "balance_loss_clip": 1.02257693,
      "balance_loss_mlp": 1.02051926,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 2.300406281975205,
      "language_loss": 0.79664963,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.81912065,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 3.6257827281951904
    },
    {
      "auxiliary_loss_clip": 0.01209743,
      "auxiliary_loss_mlp": 0.01041204,
      "balance_loss_clip": 0.98081303,
      "balance_loss_mlp": 1.03187037,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 1.7977168192263553,
      "language_loss": 0.8767978,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.89930725,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 2.706766366958618
    },
    {
      "auxiliary_loss_clip": 0.01210902,
      "auxiliary_loss_mlp": 0.01034937,
      "balance_loss_clip": 1.02249825,
      "balance_loss_mlp": 1.02482176,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 2.0888412604270044,
      "language_loss": 0.84380221,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86626065,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 3.6050846576690674
    },
    {
      "auxiliary_loss_clip": 0.01216104,
      "auxiliary_loss_mlp": 0.01044076,
      "balance_loss_clip": 1.02581632,
      "balance_loss_mlp": 1.03340721,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 1.8511063995963524,
      "language_loss": 0.76987898,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.79248083,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 3.5428855419158936
    },
    {
      "auxiliary_loss_clip": 0.01209376,
      "auxiliary_loss_mlp": 0.01034054,
      "balance_loss_clip": 1.02127981,
      "balance_loss_mlp": 1.02356339,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 1.8131444054927452,
      "language_loss": 0.70467103,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.72710538,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.6247663497924805
    },
    {
      "auxiliary_loss_clip": 0.01111009,
      "auxiliary_loss_mlp": 0.01003805,
      "balance_loss_clip": 0.87285703,
      "balance_loss_mlp": 0.99948943,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.7915168599982011,
      "language_loss": 0.64825463,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66940272,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.3941307067871094
    },
    {
      "auxiliary_loss_clip": 0.01111763,
      "auxiliary_loss_mlp": 0.01122326,
      "balance_loss_clip": 0.91055906,
      "balance_loss_mlp": 0.0,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8625915952141087,
      "language_loss": 0.6292243,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.65156519,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.2578420639038086
    },
    {
      "auxiliary_loss_clip": 0.01201948,
      "auxiliary_loss_mlp": 0.01039968,
      "balance_loss_clip": 0.90467274,
      "balance_loss_mlp": 1.02989507,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 1.7594934875151573,
      "language_loss": 0.73077041,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75318956,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.7543931007385254
    },
    {
      "auxiliary_loss_clip": 0.01216129,
      "auxiliary_loss_mlp": 0.01038921,
      "balance_loss_clip": 0.98355681,
      "balance_loss_mlp": 1.02908659,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.287716928285902,
      "language_loss": 0.80691826,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.82946879,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.7486226558685303
    },
    {
      "auxiliary_loss_clip": 0.01207235,
      "auxiliary_loss_mlp": 0.01043422,
      "balance_loss_clip": 0.98328102,
      "balance_loss_mlp": 1.03273475,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 1.8355548491332054,
      "language_loss": 0.8453815,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86788809,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.7395660877227783
    },
    {
      "auxiliary_loss_clip": 0.01108763,
      "auxiliary_loss_mlp": 0.01003162,
      "balance_loss_clip": 0.95172405,
      "balance_loss_mlp": 0.99929965,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.9180039269587318,
      "language_loss": 0.64228964,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66340899,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.136599063873291
    },
    {
      "auxiliary_loss_clip": 0.0121173,
      "auxiliary_loss_mlp": 0.01038303,
      "balance_loss_clip": 1.02311599,
      "balance_loss_mlp": 1.02755606,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 4.209610341129893,
      "language_loss": 0.78512824,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80762857,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.689096689224243
    },
    {
      "auxiliary_loss_clip": 0.01221497,
      "auxiliary_loss_mlp": 0.0104369,
      "balance_loss_clip": 1.06629217,
      "balance_loss_mlp": 1.03237104,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.123861814382286,
      "language_loss": 0.72649068,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.74914253,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.669870138168335
    },
    {
      "auxiliary_loss_clip": 0.01110285,
      "auxiliary_loss_mlp": 0.01004261,
      "balance_loss_clip": 0.98855996,
      "balance_loss_mlp": 1.00025606,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.8361843826368187,
      "language_loss": 0.64860541,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.66975087,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.2534613609313965
    },
    {
      "auxiliary_loss_clip": 0.01214313,
      "auxiliary_loss_mlp": 0.01038569,
      "balance_loss_clip": 1.06134009,
      "balance_loss_mlp": 1.02784598,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 2.120970771874738,
      "language_loss": 0.87565923,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.89818799,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 2.6074204444885254
    },
    {
      "auxiliary_loss_clip": 0.0121461,
      "auxiliary_loss_mlp": 0.01035562,
      "balance_loss_clip": 1.0613035,
      "balance_loss_mlp": 1.02570963,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 1.7905733447391534,
      "language_loss": 0.74518585,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.76768756,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.671447515487671
    },
    {
      "auxiliary_loss_clip": 0.01212552,
      "auxiliary_loss_mlp": 0.01040461,
      "balance_loss_clip": 1.02379131,
      "balance_loss_mlp": 1.03117454,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 1.8832105609634828,
      "language_loss": 0.77469903,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.79722917,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.6968953609466553
    },
    {
      "auxiliary_loss_clip": 0.01216902,
      "auxiliary_loss_mlp": 0.01033516,
      "balance_loss_clip": 1.05949819,
      "balance_loss_mlp": 1.02266181,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 2.5621541135228245,
      "language_loss": 0.76885426,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.79135847,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.646167755126953
    },
    {
      "auxiliary_loss_clip": 0.01205184,
      "auxiliary_loss_mlp": 0.01031048,
      "balance_loss_clip": 0.86976767,
      "balance_loss_mlp": 1.02092671,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 2.3177165365970085,
      "language_loss": 0.84775156,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87011385,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.843153715133667
    },
    {
      "auxiliary_loss_clip": 0.01218797,
      "auxiliary_loss_mlp": 0.01041796,
      "balance_loss_clip": 1.022228,
      "balance_loss_mlp": 1.0316695,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 1.6557264455502965,
      "language_loss": 0.9129039,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.9355098,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 2.819718837738037
    },
    {
      "auxiliary_loss_clip": 0.01107229,
      "auxiliary_loss_mlp": 0.01007136,
      "balance_loss_clip": 0.94934583,
      "balance_loss_mlp": 1.00320232,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.7495482510009539,
      "language_loss": 0.59409618,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61523986,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 3.391648530960083
    },
    {
      "auxiliary_loss_clip": 0.01214515,
      "auxiliary_loss_mlp": 0.01036038,
      "balance_loss_clip": 0.94838965,
      "balance_loss_mlp": 1.02526712,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 2.3890903830609607,
      "language_loss": 0.80474883,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.8272543,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.7522835731506348
    },
    {
      "auxiliary_loss_clip": 0.0121581,
      "auxiliary_loss_mlp": 0.0104594,
      "balance_loss_clip": 1.02124548,
      "balance_loss_mlp": 1.03483534,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 2.2970508950221,
      "language_loss": 0.83488172,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.85749924,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 2.6767585277557373
    },
    {
      "auxiliary_loss_clip": 0.01216932,
      "auxiliary_loss_mlp": 0.01041521,
      "balance_loss_clip": 1.06387258,
      "balance_loss_mlp": 1.03103602,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.15434159653544,
      "language_loss": 0.80936754,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.83195198,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.6340324878692627
    },
    {
      "auxiliary_loss_clip": 0.01219083,
      "auxiliary_loss_mlp": 0.01034856,
      "balance_loss_clip": 1.02659619,
      "balance_loss_mlp": 1.02483678,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 1.9418885802737305,
      "language_loss": 0.81447983,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83701921,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 3.6209588050842285
    },
    {
      "auxiliary_loss_clip": 0.01214624,
      "auxiliary_loss_mlp": 0.01035711,
      "balance_loss_clip": 0.98528099,
      "balance_loss_mlp": 1.02549529,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 2.7397030302524934,
      "language_loss": 0.78164363,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.80414701,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 2.6929092407226562
    },
    {
      "auxiliary_loss_clip": 0.01100331,
      "auxiliary_loss_mlp": 0.01002839,
      "balance_loss_clip": 0.94505513,
      "balance_loss_mlp": 0.99864239,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.7951995229715979,
      "language_loss": 0.54248869,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56352043,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 3.2231850624084473
    },
    {
      "auxiliary_loss_clip": 0.01220507,
      "auxiliary_loss_mlp": 0.01048861,
      "balance_loss_clip": 1.06556368,
      "balance_loss_mlp": 1.03871,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 1.8950053797454351,
      "language_loss": 0.77210665,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.79480028,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 3.6082534790039062
    },
    {
      "auxiliary_loss_clip": 0.01208453,
      "auxiliary_loss_mlp": 0.01035946,
      "balance_loss_clip": 0.98616678,
      "balance_loss_mlp": 1.02574146,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 1.8000883517938093,
      "language_loss": 0.81382006,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83626401,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 3.676081657409668
    },
    {
      "auxiliary_loss_clip": 0.01197511,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 0.90662307,
      "balance_loss_mlp": 1.01922894,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.6655924494788856,
      "language_loss": 0.73133039,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.7536006,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 2.911912441253662
    },
    {
      "auxiliary_loss_clip": 0.01097285,
      "auxiliary_loss_mlp": 0.01006448,
      "balance_loss_clip": 0.91357875,
      "balance_loss_mlp": 1.00246608,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9052104968799436,
      "language_loss": 0.58228505,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60332239,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 4.1792333126068115
    },
    {
      "auxiliary_loss_clip": 0.01218856,
      "auxiliary_loss_mlp": 0.01037123,
      "balance_loss_clip": 0.98439419,
      "balance_loss_mlp": 1.02665687,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 1.5246830671181215,
      "language_loss": 0.80450225,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.82706201,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.7312982082366943
    },
    {
      "auxiliary_loss_clip": 0.0121218,
      "auxiliary_loss_mlp": 0.01042787,
      "balance_loss_clip": 1.06074178,
      "balance_loss_mlp": 1.03294039,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 2.765690762724468,
      "language_loss": 0.86725801,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.8898077,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.713315486907959
    },
    {
      "auxiliary_loss_clip": 0.01218806,
      "auxiliary_loss_mlp": 0.01043648,
      "balance_loss_clip": 1.06365716,
      "balance_loss_mlp": 1.03333092,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 1.9350444947636942,
      "language_loss": 0.71427858,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73690313,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.6702818870544434
    },
    {
      "auxiliary_loss_clip": 0.01204979,
      "auxiliary_loss_mlp": 0.01037758,
      "balance_loss_clip": 0.98156762,
      "balance_loss_mlp": 1.02742279,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 1.5297989414576265,
      "language_loss": 0.84487468,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.86730206,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.867764472961426
    },
    {
      "auxiliary_loss_clip": 0.01200244,
      "auxiliary_loss_mlp": 0.01038612,
      "balance_loss_clip": 0.94162434,
      "balance_loss_mlp": 1.02779388,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 2.0825959887440146,
      "language_loss": 0.73046571,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.75285423,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.8377954959869385
    },
    {
      "auxiliary_loss_clip": 0.01210946,
      "auxiliary_loss_mlp": 0.01040425,
      "balance_loss_clip": 0.94361222,
      "balance_loss_mlp": 1.03094184,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 1.7225063150212934,
      "language_loss": 0.8405844,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86309808,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.764893054962158
    },
    {
      "auxiliary_loss_clip": 0.01216139,
      "auxiliary_loss_mlp": 0.01034645,
      "balance_loss_clip": 1.02089596,
      "balance_loss_mlp": 1.02425003,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 1.9926998741013027,
      "language_loss": 0.74906623,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.77157414,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.699862003326416
    },
    {
      "auxiliary_loss_clip": 0.01210525,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 1.02199161,
      "balance_loss_mlp": 1.02559304,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 1.977545446063583,
      "language_loss": 0.77453309,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.79699463,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 2.6282689571380615
    },
    {
      "auxiliary_loss_clip": 0.01208689,
      "auxiliary_loss_mlp": 0.0104047,
      "balance_loss_clip": 0.94624209,
      "balance_loss_mlp": 1.0304023,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 1.9793527487072846,
      "language_loss": 0.78422773,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.80671936,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 2.7714149951934814
    },
    {
      "auxiliary_loss_clip": 0.01186727,
      "auxiliary_loss_mlp": 0.01036571,
      "balance_loss_clip": 0.90223694,
      "balance_loss_mlp": 1.0260036,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 1.825043157382343,
      "language_loss": 0.77403158,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79626453,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.749321460723877
    },
    {
      "auxiliary_loss_clip": 0.01197897,
      "auxiliary_loss_mlp": 0.01036766,
      "balance_loss_clip": 0.98180401,
      "balance_loss_mlp": 1.02570343,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 3.009256429861296,
      "language_loss": 0.76522708,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.7875737,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.755347728729248
    },
    {
      "auxiliary_loss_clip": 0.0121107,
      "auxiliary_loss_mlp": 0.01032439,
      "balance_loss_clip": 1.02326441,
      "balance_loss_mlp": 1.022789,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 2.3706007812011665,
      "language_loss": 0.92091715,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94335222,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.711890697479248
    },
    {
      "auxiliary_loss_clip": 0.01212414,
      "auxiliary_loss_mlp": 0.01038892,
      "balance_loss_clip": 0.94735098,
      "balance_loss_mlp": 1.02930176,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 3.0819686846166374,
      "language_loss": 0.7485857,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.77109885,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 2.6879703998565674
    },
    {
      "auxiliary_loss_clip": 0.01210446,
      "auxiliary_loss_mlp": 0.01035674,
      "balance_loss_clip": 0.97998536,
      "balance_loss_mlp": 1.02567816,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 1.7246550135791279,
      "language_loss": 0.85913527,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.88159645,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.7534260749816895
    },
    {
      "auxiliary_loss_clip": 0.01207896,
      "auxiliary_loss_mlp": 0.01034483,
      "balance_loss_clip": 0.98708189,
      "balance_loss_mlp": 1.02473772,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 2.1257968753989487,
      "language_loss": 0.8903116,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.9127354,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.6645336151123047
    },
    {
      "auxiliary_loss_clip": 0.01211395,
      "auxiliary_loss_mlp": 0.01034839,
      "balance_loss_clip": 1.06066084,
      "balance_loss_mlp": 1.02460432,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.0900391484499163,
      "language_loss": 0.81867445,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.84113675,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 2.5995967388153076
    },
    {
      "auxiliary_loss_clip": 0.01198723,
      "auxiliary_loss_mlp": 0.01037215,
      "balance_loss_clip": 0.78847426,
      "balance_loss_mlp": 1.02750516,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 1.9655895647734132,
      "language_loss": 0.71239316,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.7347526,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 3.152052402496338
    },
    {
      "auxiliary_loss_clip": 0.01190255,
      "auxiliary_loss_mlp": 0.01039706,
      "balance_loss_clip": 0.97874486,
      "balance_loss_mlp": 1.02965629,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.0537167144046293,
      "language_loss": 0.89826846,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.92056805,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 3.066774368286133
    },
    {
      "auxiliary_loss_clip": 0.0109885,
      "auxiliary_loss_mlp": 0.01008437,
      "balance_loss_clip": 0.94438267,
      "balance_loss_mlp": 1.00419283,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.7596701838641039,
      "language_loss": 0.55527431,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57634723,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.2797293663024902
    },
    {
      "auxiliary_loss_clip": 0.01198294,
      "auxiliary_loss_mlp": 0.01039314,
      "balance_loss_clip": 0.98054564,
      "balance_loss_mlp": 1.02869248,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 2.8112759933013343,
      "language_loss": 0.81198716,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.83436322,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 4.07315468788147
    },
    {
      "auxiliary_loss_clip": 0.01202748,
      "auxiliary_loss_mlp": 0.01029494,
      "balance_loss_clip": 0.90816832,
      "balance_loss_mlp": 1.01961803,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.1271894991696016,
      "language_loss": 0.74972361,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77204609,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 2.865281820297241
    },
    {
      "auxiliary_loss_clip": 0.0121597,
      "auxiliary_loss_mlp": 0.0112696,
      "balance_loss_clip": 1.02832878,
      "balance_loss_mlp": 0.0,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 1.9395984686344452,
      "language_loss": 0.84725404,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.87068337,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 2.6714277267456055
    },
    {
      "auxiliary_loss_clip": 0.01199957,
      "auxiliary_loss_mlp": 0.01032394,
      "balance_loss_clip": 0.9789086,
      "balance_loss_mlp": 1.02212977,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 1.8305796436490667,
      "language_loss": 0.78857219,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81089562,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 3.696798086166382
    },
    {
      "auxiliary_loss_clip": 0.01205499,
      "auxiliary_loss_mlp": 0.01039664,
      "balance_loss_clip": 0.98288053,
      "balance_loss_mlp": 1.03017461,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 2.1560942440523347,
      "language_loss": 0.80629176,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82874346,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 3.5720059871673584
    },
    {
      "auxiliary_loss_clip": 0.0119963,
      "auxiliary_loss_mlp": 0.0102631,
      "balance_loss_clip": 1.02010441,
      "balance_loss_mlp": 1.01634455,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 3.5146576091333572,
      "language_loss": 0.81755161,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.83981109,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 2.6988625526428223
    },
    {
      "auxiliary_loss_clip": 0.01211162,
      "auxiliary_loss_mlp": 0.01035226,
      "balance_loss_clip": 1.02358675,
      "balance_loss_mlp": 1.02557039,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 1.627994926204797,
      "language_loss": 0.72862756,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.75109148,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.742434501647949
    },
    {
      "auxiliary_loss_clip": 0.01217736,
      "auxiliary_loss_mlp": 0.01032463,
      "balance_loss_clip": 1.06266975,
      "balance_loss_mlp": 1.02207935,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 2.130592004755542,
      "language_loss": 0.81446749,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83696949,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 3.5448403358459473
    },
    {
      "auxiliary_loss_clip": 0.01208736,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.01852369,
      "balance_loss_mlp": 1.01986074,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 2.501636664849308,
      "language_loss": 0.69176042,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.71414495,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.6697030067443848
    },
    {
      "auxiliary_loss_clip": 0.01213178,
      "auxiliary_loss_mlp": 0.01037111,
      "balance_loss_clip": 1.02102947,
      "balance_loss_mlp": 1.02772307,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.5559506759633706,
      "language_loss": 0.86486912,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.88737196,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.713487148284912
    },
    {
      "auxiliary_loss_clip": 0.01205862,
      "auxiliary_loss_mlp": 0.01037455,
      "balance_loss_clip": 1.02107978,
      "balance_loss_mlp": 1.02826989,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 1.5901028851714656,
      "language_loss": 0.73373502,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.75616825,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.922863721847534
    },
    {
      "auxiliary_loss_clip": 0.01191471,
      "auxiliary_loss_mlp": 0.0103865,
      "balance_loss_clip": 0.97710896,
      "balance_loss_mlp": 1.02877319,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 1.787545677851132,
      "language_loss": 0.78545767,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80775887,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.7186625003814697
    },
    {
      "auxiliary_loss_clip": 0.01198703,
      "auxiliary_loss_mlp": 0.0103743,
      "balance_loss_clip": 0.97991407,
      "balance_loss_mlp": 1.02730954,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 1.7828528873404648,
      "language_loss": 0.85655725,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.87891853,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.7628142833709717
    },
    {
      "auxiliary_loss_clip": 0.01208653,
      "auxiliary_loss_mlp": 0.01038454,
      "balance_loss_clip": 1.02199316,
      "balance_loss_mlp": 1.02853572,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 1.8540477690132828,
      "language_loss": 0.85432911,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87680018,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.670884609222412
    },
    {
      "auxiliary_loss_clip": 0.01201314,
      "auxiliary_loss_mlp": 0.01034094,
      "balance_loss_clip": 0.98050117,
      "balance_loss_mlp": 1.02452159,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 2.2746366363089567,
      "language_loss": 0.60115147,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.62350553,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.7545957565307617
    },
    {
      "auxiliary_loss_clip": 0.012034,
      "auxiliary_loss_mlp": 0.01036198,
      "balance_loss_clip": 0.94254684,
      "balance_loss_mlp": 1.02706099,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.23688694275574,
      "language_loss": 0.91885614,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.94125217,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.7750234603881836
    },
    {
      "auxiliary_loss_clip": 0.01213941,
      "auxiliary_loss_mlp": 0.01034284,
      "balance_loss_clip": 0.98179376,
      "balance_loss_mlp": 1.02446747,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 1.9125493953461077,
      "language_loss": 0.71481812,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.7373004,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.7321228981018066
    },
    {
      "auxiliary_loss_clip": 0.01194564,
      "auxiliary_loss_mlp": 0.0103143,
      "balance_loss_clip": 0.98031902,
      "balance_loss_mlp": 1.02171421,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 2.1465886436509583,
      "language_loss": 0.83331645,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.8555764,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.6529908180236816
    },
    {
      "auxiliary_loss_clip": 0.01197707,
      "auxiliary_loss_mlp": 0.01032625,
      "balance_loss_clip": 0.98154098,
      "balance_loss_mlp": 1.02261782,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 1.6420502750304486,
      "language_loss": 0.73463786,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.75694114,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.695159912109375
    },
    {
      "auxiliary_loss_clip": 0.01209184,
      "auxiliary_loss_mlp": 0.01032786,
      "balance_loss_clip": 0.94486439,
      "balance_loss_mlp": 1.02312982,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 1.828577287306282,
      "language_loss": 0.86565638,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.88807607,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 2.6791460514068604
    },
    {
      "auxiliary_loss_clip": 0.012101,
      "auxiliary_loss_mlp": 0.01034502,
      "balance_loss_clip": 1.05979657,
      "balance_loss_mlp": 1.02498293,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 1.8135752296191536,
      "language_loss": 0.92198318,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94442928,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.6057419776916504
    },
    {
      "auxiliary_loss_clip": 0.01215656,
      "auxiliary_loss_mlp": 0.01044097,
      "balance_loss_clip": 0.9853459,
      "balance_loss_mlp": 1.03350496,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 2.8967518321344383,
      "language_loss": 0.73489058,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.75748813,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.6895735263824463
    },
    {
      "auxiliary_loss_clip": 0.01205944,
      "auxiliary_loss_mlp": 0.01034686,
      "balance_loss_clip": 0.98146367,
      "balance_loss_mlp": 1.02499986,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 1.8029577927204383,
      "language_loss": 0.88281524,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90522152,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 2.6458144187927246
    },
    {
      "auxiliary_loss_clip": 0.01201369,
      "auxiliary_loss_mlp": 0.0103502,
      "balance_loss_clip": 0.94615757,
      "balance_loss_mlp": 1.0248158,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 1.7574844890329484,
      "language_loss": 0.76758689,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78995079,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 2.7060909271240234
    },
    {
      "auxiliary_loss_clip": 0.01211377,
      "auxiliary_loss_mlp": 0.01126273,
      "balance_loss_clip": 1.06119204,
      "balance_loss_mlp": 0.0,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.75728107040267,
      "language_loss": 0.8046875,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.82806396,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 2.638603448867798
    },
    {
      "auxiliary_loss_clip": 0.0120617,
      "auxiliary_loss_mlp": 0.01030917,
      "balance_loss_clip": 0.98134792,
      "balance_loss_mlp": 1.02083182,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 2.0403266791622325,
      "language_loss": 0.69849443,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.72086537,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 2.7678043842315674
    },
    {
      "auxiliary_loss_clip": 0.01210643,
      "auxiliary_loss_mlp": 0.01036578,
      "balance_loss_clip": 0.94428951,
      "balance_loss_mlp": 1.02634358,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 1.8331229259817032,
      "language_loss": 0.81738806,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.83986026,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 3.701821804046631
    },
    {
      "auxiliary_loss_clip": 0.01210238,
      "auxiliary_loss_mlp": 0.01033087,
      "balance_loss_clip": 1.01913321,
      "balance_loss_mlp": 1.02312708,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 2.8857030432455435,
      "language_loss": 0.7938754,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.81630868,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 2.73474383354187
    },
    {
      "auxiliary_loss_clip": 0.01210972,
      "auxiliary_loss_mlp": 0.01036457,
      "balance_loss_clip": 1.02064061,
      "balance_loss_mlp": 1.02731943,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.5587543232498293,
      "language_loss": 0.74564958,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76812387,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.6941936016082764
    },
    {
      "auxiliary_loss_clip": 0.01207531,
      "auxiliary_loss_mlp": 0.0104013,
      "balance_loss_clip": 1.05856431,
      "balance_loss_mlp": 1.0314697,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.551897521039097,
      "language_loss": 0.8035745,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82605112,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 3.6249213218688965
    },
    {
      "auxiliary_loss_clip": 0.01107333,
      "auxiliary_loss_mlp": 0.01005951,
      "balance_loss_clip": 0.95319927,
      "balance_loss_mlp": 1.00199366,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.8647442597834171,
      "language_loss": 0.57880497,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.5999378,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.297166347503662
    },
    {
      "auxiliary_loss_clip": 0.01206633,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 0.94384176,
      "balance_loss_mlp": 1.02118993,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 2.2955765856366064,
      "language_loss": 0.67466295,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.69703376,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 3.6210803985595703
    },
    {
      "auxiliary_loss_clip": 0.01210685,
      "auxiliary_loss_mlp": 0.01036082,
      "balance_loss_clip": 1.06120741,
      "balance_loss_mlp": 1.02693319,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 2.4129168521884568,
      "language_loss": 0.74431872,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.7667864,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 3.5643422603607178
    },
    {
      "auxiliary_loss_clip": 0.01216796,
      "auxiliary_loss_mlp": 0.01033012,
      "balance_loss_clip": 1.06167591,
      "balance_loss_mlp": 1.02294517,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 1.7636364025984257,
      "language_loss": 0.73274374,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.75524187,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.6954214572906494
    },
    {
      "auxiliary_loss_clip": 0.01107995,
      "auxiliary_loss_mlp": 0.01003512,
      "balance_loss_clip": 0.99348986,
      "balance_loss_mlp": 0.99948233,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8518779138210567,
      "language_loss": 0.58405733,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.6051724,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.298980474472046
    },
    {
      "auxiliary_loss_clip": 0.01208419,
      "auxiliary_loss_mlp": 0.01031726,
      "balance_loss_clip": 1.05870807,
      "balance_loss_mlp": 1.02192068,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 1.7627973998841115,
      "language_loss": 0.85664028,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.87904167,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.6340091228485107
    },
    {
      "auxiliary_loss_clip": 0.01200336,
      "auxiliary_loss_mlp": 0.01037281,
      "balance_loss_clip": 0.94101286,
      "balance_loss_mlp": 1.02785802,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.1440648570078333,
      "language_loss": 0.76220375,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78457993,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.8141074180603027
    },
    {
      "auxiliary_loss_clip": 0.01212534,
      "auxiliary_loss_mlp": 0.01034124,
      "balance_loss_clip": 1.06013918,
      "balance_loss_mlp": 1.02462947,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.56691203657703,
      "language_loss": 0.77116048,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79362702,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.6768457889556885
    },
    {
      "auxiliary_loss_clip": 0.01215215,
      "auxiliary_loss_mlp": 0.01127278,
      "balance_loss_clip": 1.02018106,
      "balance_loss_mlp": 0.0,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 1.9444290995809281,
      "language_loss": 0.82411194,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.84753692,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.6388957500457764
    },
    {
      "auxiliary_loss_clip": 0.01202177,
      "auxiliary_loss_mlp": 0.0103163,
      "balance_loss_clip": 0.93881279,
      "balance_loss_mlp": 1.02097321,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 2.6475035324173253,
      "language_loss": 0.76467484,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.78701288,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.721461534500122
    },
    {
      "auxiliary_loss_clip": 0.01209874,
      "auxiliary_loss_mlp": 0.01034756,
      "balance_loss_clip": 1.05733514,
      "balance_loss_mlp": 1.02511787,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 1.9605441783979554,
      "language_loss": 0.70075393,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.7232002,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.733931303024292
    },
    {
      "auxiliary_loss_clip": 0.01200876,
      "auxiliary_loss_mlp": 0.01034742,
      "balance_loss_clip": 1.02041268,
      "balance_loss_mlp": 1.02564049,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 2.1228804200429066,
      "language_loss": 0.89145905,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91381526,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.61442232131958
    },
    {
      "auxiliary_loss_clip": 0.01201823,
      "auxiliary_loss_mlp": 0.01032542,
      "balance_loss_clip": 0.98317212,
      "balance_loss_mlp": 1.0231539,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 1.846370980521908,
      "language_loss": 0.76861405,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.79095781,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.690269947052002
    },
    {
      "auxiliary_loss_clip": 0.01198274,
      "auxiliary_loss_mlp": 0.01031,
      "balance_loss_clip": 0.90572888,
      "balance_loss_mlp": 1.02183867,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.7005229654472211,
      "language_loss": 0.87061244,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89290524,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.707629919052124
    },
    {
      "auxiliary_loss_clip": 0.01205355,
      "auxiliary_loss_mlp": 0.01033384,
      "balance_loss_clip": 0.98399633,
      "balance_loss_mlp": 1.02377021,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 1.9245072465162634,
      "language_loss": 0.78631723,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.80870461,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 2.72959041595459
    },
    {
      "auxiliary_loss_clip": 0.01205422,
      "auxiliary_loss_mlp": 0.01126853,
      "balance_loss_clip": 0.98594809,
      "balance_loss_mlp": 0.0,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 2.0975783418912166,
      "language_loss": 0.89308393,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.91640675,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.705643653869629
    },
    {
      "auxiliary_loss_clip": 0.0121252,
      "auxiliary_loss_mlp": 0.01032104,
      "balance_loss_clip": 1.02064228,
      "balance_loss_mlp": 1.02263927,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 2.2049693942747197,
      "language_loss": 0.78396034,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.80640662,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 2.646941661834717
    },
    {
      "auxiliary_loss_clip": 0.01207029,
      "auxiliary_loss_mlp": 0.0103786,
      "balance_loss_clip": 0.98173881,
      "balance_loss_mlp": 1.02828729,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 3.777917579477249,
      "language_loss": 0.70889181,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.73134071,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.6871087551116943
    },
    {
      "auxiliary_loss_clip": 0.01212056,
      "auxiliary_loss_mlp": 0.01036657,
      "balance_loss_clip": 1.06129503,
      "balance_loss_mlp": 1.02712619,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.1725419078712966,
      "language_loss": 0.80704224,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.82952935,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 2.585860252380371
    },
    {
      "auxiliary_loss_clip": 0.0119388,
      "auxiliary_loss_mlp": 0.01126935,
      "balance_loss_clip": 0.94365597,
      "balance_loss_mlp": 0.0,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 1.9484225340322876,
      "language_loss": 0.83773017,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.86093837,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.666769504547119
    },
    {
      "auxiliary_loss_clip": 0.01210843,
      "auxiliary_loss_mlp": 0.01032186,
      "balance_loss_clip": 1.01952219,
      "balance_loss_mlp": 1.02228618,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 1.5852778228094264,
      "language_loss": 0.68374717,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.70617747,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 2.6625890731811523
    },
    {
      "auxiliary_loss_clip": 0.01212759,
      "auxiliary_loss_mlp": 0.01032738,
      "balance_loss_clip": 1.06158638,
      "balance_loss_mlp": 1.02281392,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 1.7374880014119651,
      "language_loss": 0.66494405,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.68739903,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 2.5661141872406006
    },
    {
      "auxiliary_loss_clip": 0.01207909,
      "auxiliary_loss_mlp": 0.01031892,
      "balance_loss_clip": 0.90923107,
      "balance_loss_mlp": 1.02221179,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 1.89471125420174,
      "language_loss": 0.85455143,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.87694943,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 2.757378339767456
    },
    {
      "auxiliary_loss_clip": 0.01202676,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 0.9813295,
      "balance_loss_mlp": 1.02121162,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.916313140511152,
      "language_loss": 0.79589158,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.81822312,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 3.618104934692383
    },
    {
      "auxiliary_loss_clip": 0.01204445,
      "auxiliary_loss_mlp": 0.01034024,
      "balance_loss_clip": 0.98472303,
      "balance_loss_mlp": 1.02414787,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 2.7170567730372213,
      "language_loss": 0.82348138,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.84586608,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.6277949810028076
    },
    {
      "auxiliary_loss_clip": 0.01214153,
      "auxiliary_loss_mlp": 0.01031933,
      "balance_loss_clip": 0.98577142,
      "balance_loss_mlp": 1.02219963,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.1357994395246713,
      "language_loss": 0.69058406,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71304494,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 3.6749680042266846
    },
    {
      "auxiliary_loss_clip": 0.01195085,
      "auxiliary_loss_mlp": 0.01029736,
      "balance_loss_clip": 0.90141469,
      "balance_loss_mlp": 1.01978791,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 1.7332193811612189,
      "language_loss": 0.74163592,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.76388407,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 3.64876651763916
    },
    {
      "auxiliary_loss_clip": 0.01207635,
      "auxiliary_loss_mlp": 0.01036718,
      "balance_loss_clip": 0.90382409,
      "balance_loss_mlp": 1.02641881,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 1.8941887931808712,
      "language_loss": 0.80497599,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.82741952,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 2.705787181854248
    },
    {
      "auxiliary_loss_clip": 0.01098459,
      "auxiliary_loss_mlp": 0.01005236,
      "balance_loss_clip": 0.98472667,
      "balance_loss_mlp": 1.00106323,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8437609731874023,
      "language_loss": 0.57933718,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.60037416,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 4.1306281089782715
    },
    {
      "auxiliary_loss_clip": 0.01199572,
      "auxiliary_loss_mlp": 0.01035614,
      "balance_loss_clip": 1.01733947,
      "balance_loss_mlp": 1.02579117,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 2.4106558020972835,
      "language_loss": 0.83204049,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85439235,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.721031665802002
    },
    {
      "auxiliary_loss_clip": 0.01203742,
      "auxiliary_loss_mlp": 0.01032025,
      "balance_loss_clip": 0.9452405,
      "balance_loss_mlp": 1.02263784,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 1.8377323935345544,
      "language_loss": 0.80644834,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82880598,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.7184345722198486
    },
    {
      "auxiliary_loss_clip": 0.01210632,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 0.98358858,
      "balance_loss_mlp": 1.01926756,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 2.0014161580785204,
      "language_loss": 0.78217226,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.80456567,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.6809468269348145
    },
    {
      "auxiliary_loss_clip": 0.01205828,
      "auxiliary_loss_mlp": 0.01038519,
      "balance_loss_clip": 0.94362271,
      "balance_loss_mlp": 1.02859497,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.2157647230109676,
      "language_loss": 0.79591978,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.81836331,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.682063341140747
    },
    {
      "auxiliary_loss_clip": 0.01212795,
      "auxiliary_loss_mlp": 0.01030699,
      "balance_loss_clip": 1.06065404,
      "balance_loss_mlp": 1.02044094,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 1.797257743130324,
      "language_loss": 0.86437595,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88681084,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.715782403945923
    },
    {
      "auxiliary_loss_clip": 0.01206517,
      "auxiliary_loss_mlp": 0.01127082,
      "balance_loss_clip": 0.98493695,
      "balance_loss_mlp": 0.0,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 2.19755403446841,
      "language_loss": 0.89060664,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.91394269,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.656130075454712
    },
    {
      "auxiliary_loss_clip": 0.01205325,
      "auxiliary_loss_mlp": 0.01036325,
      "balance_loss_clip": 0.94532061,
      "balance_loss_mlp": 1.02614462,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.042177400651882,
      "language_loss": 0.90388811,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92630458,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.7583799362182617
    },
    {
      "auxiliary_loss_clip": 0.01199161,
      "auxiliary_loss_mlp": 0.01038923,
      "balance_loss_clip": 0.94220805,
      "balance_loss_mlp": 1.02879047,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 2.4076661779083843,
      "language_loss": 0.86778963,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.89017045,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.6498680114746094
    },
    {
      "auxiliary_loss_clip": 0.01210583,
      "auxiliary_loss_mlp": 0.01039554,
      "balance_loss_clip": 0.98160452,
      "balance_loss_mlp": 1.02927208,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 2.1596474975592215,
      "language_loss": 0.89473492,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91723627,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.7770700454711914
    },
    {
      "auxiliary_loss_clip": 0.01212009,
      "auxiliary_loss_mlp": 0.01035891,
      "balance_loss_clip": 1.02202106,
      "balance_loss_mlp": 1.02542424,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 1.8962555513492712,
      "language_loss": 0.7941131,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.8165921,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.756913661956787
    },
    {
      "auxiliary_loss_clip": 0.0120711,
      "auxiliary_loss_mlp": 0.01034775,
      "balance_loss_clip": 0.90300494,
      "balance_loss_mlp": 1.02409935,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 3.307973537047587,
      "language_loss": 0.74971604,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.7721349,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.943244218826294
    },
    {
      "auxiliary_loss_clip": 0.01213449,
      "auxiliary_loss_mlp": 0.01045654,
      "balance_loss_clip": 1.02147281,
      "balance_loss_mlp": 1.0347048,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 2.2206375469338306,
      "language_loss": 0.84246993,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.86506104,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.7755966186523438
    },
    {
      "auxiliary_loss_clip": 0.01210606,
      "auxiliary_loss_mlp": 0.0103374,
      "balance_loss_clip": 1.02172947,
      "balance_loss_mlp": 1.0238874,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 1.8873259615048978,
      "language_loss": 0.83347249,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.8559159,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 2.6802282333374023
    },
    {
      "auxiliary_loss_clip": 0.01212662,
      "auxiliary_loss_mlp": 0.01031385,
      "balance_loss_clip": 1.05877972,
      "balance_loss_mlp": 1.02091289,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 2.1195710620846664,
      "language_loss": 0.85314071,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.8755812,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.6972367763519287
    },
    {
      "auxiliary_loss_clip": 0.01211723,
      "auxiliary_loss_mlp": 0.01036955,
      "balance_loss_clip": 1.06096363,
      "balance_loss_mlp": 1.02716756,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 1.7612513510694843,
      "language_loss": 0.7525084,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.77499521,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 2.7459871768951416
    },
    {
      "auxiliary_loss_clip": 0.01212243,
      "auxiliary_loss_mlp": 0.01032845,
      "balance_loss_clip": 1.02228725,
      "balance_loss_mlp": 1.0228194,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 2.199610950589123,
      "language_loss": 0.72109169,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74354255,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.720247745513916
    },
    {
      "auxiliary_loss_clip": 0.01199484,
      "auxiliary_loss_mlp": 0.01035974,
      "balance_loss_clip": 0.8649714,
      "balance_loss_mlp": 1.02560902,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 1.5529082809825987,
      "language_loss": 0.74759096,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.76994556,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 2.9322941303253174
    },
    {
      "auxiliary_loss_clip": 0.01215498,
      "auxiliary_loss_mlp": 0.0112787,
      "balance_loss_clip": 1.06050479,
      "balance_loss_mlp": 0.0,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 1.9871525167902535,
      "language_loss": 0.87590826,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.89934188,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.7011592388153076
    },
    {
      "auxiliary_loss_clip": 0.01188884,
      "auxiliary_loss_mlp": 0.01033258,
      "balance_loss_clip": 0.82024783,
      "balance_loss_mlp": 1.0234772,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 1.7732732032346314,
      "language_loss": 0.88463855,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90685999,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 2.996454954147339
    },
    {
      "auxiliary_loss_clip": 0.0121194,
      "auxiliary_loss_mlp": 0.01028984,
      "balance_loss_clip": 1.06033754,
      "balance_loss_mlp": 1.01868439,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 1.6426917940395172,
      "language_loss": 0.7828486,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.8052578,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 2.9708003997802734
    },
    {
      "auxiliary_loss_clip": 0.01209431,
      "auxiliary_loss_mlp": 0.01036337,
      "balance_loss_clip": 1.02026367,
      "balance_loss_mlp": 1.02616882,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 4.044722657885507,
      "language_loss": 0.90652138,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.92897904,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 4.003946542739868
    },
    {
      "auxiliary_loss_clip": 0.01205209,
      "auxiliary_loss_mlp": 0.0103416,
      "balance_loss_clip": 1.01779187,
      "balance_loss_mlp": 1.02392602,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 1.734700171456285,
      "language_loss": 0.76707029,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.789464,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 3.7036006450653076
    },
    {
      "auxiliary_loss_clip": 0.01102869,
      "auxiliary_loss_mlp": 0.01005246,
      "balance_loss_clip": 0.94731772,
      "balance_loss_mlp": 1.00124061,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7512318846014371,
      "language_loss": 0.57877916,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.59986031,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.4326343536376953
    },
    {
      "auxiliary_loss_clip": 0.01209243,
      "auxiliary_loss_mlp": 0.01038674,
      "balance_loss_clip": 0.98375094,
      "balance_loss_mlp": 1.0287559,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 2.344934283533545,
      "language_loss": 0.74009287,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76257205,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 3.8713412284851074
    },
    {
      "auxiliary_loss_clip": 0.01215144,
      "auxiliary_loss_mlp": 0.01033751,
      "balance_loss_clip": 1.02334821,
      "balance_loss_mlp": 1.02384496,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 1.7336695211372315,
      "language_loss": 0.7261281,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74861705,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 2.7504992485046387
    },
    {
      "auxiliary_loss_clip": 0.01214714,
      "auxiliary_loss_mlp": 0.01037689,
      "balance_loss_clip": 1.06131959,
      "balance_loss_mlp": 1.02755022,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.058402637752043,
      "language_loss": 0.74304259,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76556659,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 3.507833957672119
    },
    {
      "auxiliary_loss_clip": 0.01203447,
      "auxiliary_loss_mlp": 0.01034136,
      "balance_loss_clip": 0.94440126,
      "balance_loss_mlp": 1.0247786,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 2.2225543535036576,
      "language_loss": 0.82104039,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.84341627,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.7942538261413574
    },
    {
      "auxiliary_loss_clip": 0.0119346,
      "auxiliary_loss_mlp": 0.01041659,
      "balance_loss_clip": 0.8617636,
      "balance_loss_mlp": 1.03078079,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 1.7736528777518572,
      "language_loss": 0.86041451,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.88276565,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.9448516368865967
    },
    {
      "auxiliary_loss_clip": 0.01206682,
      "auxiliary_loss_mlp": 0.01039632,
      "balance_loss_clip": 0.98353934,
      "balance_loss_mlp": 1.02916527,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 2.5922447509621342,
      "language_loss": 0.77191299,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79437613,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.744534492492676
    },
    {
      "auxiliary_loss_clip": 0.01197115,
      "auxiliary_loss_mlp": 0.01038727,
      "balance_loss_clip": 0.93913567,
      "balance_loss_mlp": 1.02803445,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 2.672280615006205,
      "language_loss": 0.83068585,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.85304427,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.7704854011535645
    },
    {
      "auxiliary_loss_clip": 0.01097937,
      "auxiliary_loss_mlp": 0.0100874,
      "balance_loss_clip": 1.01923919,
      "balance_loss_mlp": 1.00473464,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.7989162428947055,
      "language_loss": 0.55044663,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.57151341,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.2304675579071045
    },
    {
      "auxiliary_loss_clip": 0.01206944,
      "auxiliary_loss_mlp": 0.01035465,
      "balance_loss_clip": 0.98283648,
      "balance_loss_mlp": 1.02526665,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 2.1799924698345547,
      "language_loss": 0.72494543,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.74736953,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.7093863487243652
    },
    {
      "auxiliary_loss_clip": 0.01210593,
      "auxiliary_loss_mlp": 0.01032014,
      "balance_loss_clip": 0.94727021,
      "balance_loss_mlp": 1.02245378,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.350709544636353,
      "language_loss": 0.85747433,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87990034,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.8138840198516846
    },
    {
      "auxiliary_loss_clip": 0.01196072,
      "auxiliary_loss_mlp": 0.01037506,
      "balance_loss_clip": 0.90387845,
      "balance_loss_mlp": 1.02640724,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 1.9224111434373632,
      "language_loss": 0.72079301,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.74312878,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.817232608795166
    },
    {
      "auxiliary_loss_clip": 0.01212414,
      "auxiliary_loss_mlp": 0.01043016,
      "balance_loss_clip": 0.94222164,
      "balance_loss_mlp": 1.03303814,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 2.2380833681407637,
      "language_loss": 0.82671916,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.84927344,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.7257742881774902
    },
    {
      "auxiliary_loss_clip": 0.01095975,
      "auxiliary_loss_mlp": 0.01121873,
      "balance_loss_clip": 1.01793671,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.7579512173623888,
      "language_loss": 0.64441603,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66659451,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.134211778640747
    },
    {
      "auxiliary_loss_clip": 0.01217857,
      "auxiliary_loss_mlp": 0.01037283,
      "balance_loss_clip": 0.98493403,
      "balance_loss_mlp": 1.02748382,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 2.2415348663816017,
      "language_loss": 0.89688253,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.91943395,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.8277268409729004
    },
    {
      "auxiliary_loss_clip": 0.01211155,
      "auxiliary_loss_mlp": 0.01035132,
      "balance_loss_clip": 1.02271271,
      "balance_loss_mlp": 1.02524304,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 2.1607381834106247,
      "language_loss": 0.8184334,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.84089625,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.7314867973327637
    },
    {
      "auxiliary_loss_clip": 0.01200435,
      "auxiliary_loss_mlp": 0.01033426,
      "balance_loss_clip": 0.9792279,
      "balance_loss_mlp": 1.02377021,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 1.8383910242816262,
      "language_loss": 0.85914183,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88148046,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 2.8153669834136963
    },
    {
      "auxiliary_loss_clip": 0.01218155,
      "auxiliary_loss_mlp": 0.01033107,
      "balance_loss_clip": 1.02306533,
      "balance_loss_mlp": 1.02289701,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 2.116930449001057,
      "language_loss": 0.77715528,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.79966784,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.686563730239868
    },
    {
      "auxiliary_loss_clip": 0.01208204,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.05719185,
      "balance_loss_mlp": 1.0267874,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 1.6420343726359952,
      "language_loss": 0.75903183,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.78147769,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 2.7573282718658447
    },
    {
      "auxiliary_loss_clip": 0.01210378,
      "auxiliary_loss_mlp": 0.01036222,
      "balance_loss_clip": 0.98630714,
      "balance_loss_mlp": 1.02656007,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 1.7507122203496492,
      "language_loss": 0.75519371,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.77765965,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.7495954036712646
    },
    {
      "auxiliary_loss_clip": 0.01211569,
      "auxiliary_loss_mlp": 0.01041104,
      "balance_loss_clip": 0.9458878,
      "balance_loss_mlp": 1.03043532,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 2.077029119138681,
      "language_loss": 0.81481451,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.83734125,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.7708146572113037
    },
    {
      "auxiliary_loss_clip": 0.01204284,
      "auxiliary_loss_mlp": 0.01032676,
      "balance_loss_clip": 1.01871717,
      "balance_loss_mlp": 1.02329397,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.194179962965572,
      "language_loss": 0.81739795,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.83976758,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.641692876815796
    },
    {
      "auxiliary_loss_clip": 0.01194432,
      "auxiliary_loss_mlp": 0.01041005,
      "balance_loss_clip": 0.97891057,
      "balance_loss_mlp": 1.03139687,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 1.7336122143621808,
      "language_loss": 0.76035231,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78270662,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.6629064083099365
    },
    {
      "auxiliary_loss_clip": 0.0120785,
      "auxiliary_loss_mlp": 0.01031497,
      "balance_loss_clip": 1.02024519,
      "balance_loss_mlp": 1.02208591,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 1.6770602208456649,
      "language_loss": 0.81028503,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.8326785,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 2.68819522857666
    },
    {
      "auxiliary_loss_clip": 0.01207524,
      "auxiliary_loss_mlp": 0.0112665,
      "balance_loss_clip": 0.87041652,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.657533331850931,
      "language_loss": 0.8963623,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.91970402,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 3.685054302215576
    },
    {
      "auxiliary_loss_clip": 0.01213591,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 0.94706964,
      "balance_loss_mlp": 1.02240419,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 1.6635163363659655,
      "language_loss": 0.80076563,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82321882,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 3.7662594318389893
    },
    {
      "auxiliary_loss_clip": 0.01217895,
      "auxiliary_loss_mlp": 0.01036399,
      "balance_loss_clip": 1.02231419,
      "balance_loss_mlp": 1.02674317,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 2.1353444423697368,
      "language_loss": 0.83018672,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.85272968,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 2.772832155227661
    },
    {
      "auxiliary_loss_clip": 0.01096477,
      "auxiliary_loss_mlp": 0.01005569,
      "balance_loss_clip": 0.94342303,
      "balance_loss_mlp": 1.00168276,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7254184188327265,
      "language_loss": 0.61998963,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64101005,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 4.145323991775513
    },
    {
      "auxiliary_loss_clip": 0.01209367,
      "auxiliary_loss_mlp": 0.01038228,
      "balance_loss_clip": 0.94690299,
      "balance_loss_mlp": 1.02778554,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 1.8862779454989513,
      "language_loss": 0.75927883,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.78175479,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 2.8868303298950195
    },
    {
      "auxiliary_loss_clip": 0.01211174,
      "auxiliary_loss_mlp": 0.0103141,
      "balance_loss_clip": 1.02055609,
      "balance_loss_mlp": 1.02177179,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.1300638633757707,
      "language_loss": 0.82629037,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.8487162,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 3.6198537349700928
    },
    {
      "auxiliary_loss_clip": 0.01214156,
      "auxiliary_loss_mlp": 0.01034811,
      "balance_loss_clip": 1.02142119,
      "balance_loss_mlp": 1.02482092,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 2.34299138043537,
      "language_loss": 0.89954734,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.92203701,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.617431640625
    },
    {
      "auxiliary_loss_clip": 0.01215774,
      "auxiliary_loss_mlp": 0.01034084,
      "balance_loss_clip": 1.02182758,
      "balance_loss_mlp": 1.02342045,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 2.396182849457201,
      "language_loss": 0.73373985,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75623846,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.680549144744873
    },
    {
      "auxiliary_loss_clip": 0.01182357,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 0.89531672,
      "balance_loss_mlp": 1.02040267,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 2.5951614277909094,
      "language_loss": 0.75062215,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77274388,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.7273223400115967
    },
    {
      "auxiliary_loss_clip": 0.01210984,
      "auxiliary_loss_mlp": 0.01034656,
      "balance_loss_clip": 1.06062293,
      "balance_loss_mlp": 1.02468967,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 1.6835703658090593,
      "language_loss": 0.80820024,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.83065665,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.6425247192382812
    },
    {
      "auxiliary_loss_clip": 0.01216195,
      "auxiliary_loss_mlp": 0.01029455,
      "balance_loss_clip": 1.02330661,
      "balance_loss_mlp": 1.01962614,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 1.8576899491318886,
      "language_loss": 0.87500238,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89745891,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.6594161987304688
    },
    {
      "auxiliary_loss_clip": 0.01209959,
      "auxiliary_loss_mlp": 0.01039346,
      "balance_loss_clip": 0.98818022,
      "balance_loss_mlp": 1.03027987,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 1.7944709763312607,
      "language_loss": 0.79893613,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82142919,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.733013153076172
    },
    {
      "auxiliary_loss_clip": 0.01204865,
      "auxiliary_loss_mlp": 0.0103525,
      "balance_loss_clip": 0.98606968,
      "balance_loss_mlp": 1.0258919,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.5025234200700375,
      "language_loss": 0.79140091,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.81380212,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.8153765201568604
    },
    {
      "auxiliary_loss_clip": 0.01209449,
      "auxiliary_loss_mlp": 0.01126746,
      "balance_loss_clip": 0.9826262,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 2.3422623753109884,
      "language_loss": 0.7490654,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.77242732,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.7395637035369873
    },
    {
      "auxiliary_loss_clip": 0.01211807,
      "auxiliary_loss_mlp": 0.0103555,
      "balance_loss_clip": 1.0224762,
      "balance_loss_mlp": 1.02505398,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 1.8542066978734073,
      "language_loss": 0.80490577,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82737935,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.6658411026000977
    },
    {
      "auxiliary_loss_clip": 0.01200361,
      "auxiliary_loss_mlp": 0.01031259,
      "balance_loss_clip": 0.90418577,
      "balance_loss_mlp": 1.02169204,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 2.1530275543680757,
      "language_loss": 0.64464694,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66696316,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.745112180709839
    },
    {
      "auxiliary_loss_clip": 0.01215739,
      "auxiliary_loss_mlp": 0.01126873,
      "balance_loss_clip": 1.06433666,
      "balance_loss_mlp": 0.0,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.4852977216231333,
      "language_loss": 0.82836366,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.85178983,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.705659866333008
    },
    {
      "auxiliary_loss_clip": 0.0120841,
      "auxiliary_loss_mlp": 0.0104401,
      "balance_loss_clip": 1.02283335,
      "balance_loss_mlp": 1.03421068,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 2.449608371944514,
      "language_loss": 0.80376256,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.82628673,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.6168456077575684
    },
    {
      "auxiliary_loss_clip": 0.01192622,
      "auxiliary_loss_mlp": 0.01031851,
      "balance_loss_clip": 0.93918836,
      "balance_loss_mlp": 1.02202797,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 2.0332629703209197,
      "language_loss": 0.66557342,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.68781811,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 2.743298053741455
    },
    {
      "auxiliary_loss_clip": 0.01107515,
      "auxiliary_loss_mlp": 0.01006717,
      "balance_loss_clip": 0.94847846,
      "balance_loss_mlp": 1.00256813,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 1.6419764200436167,
      "language_loss": 0.61970216,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.64084446,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.2501697540283203
    },
    {
      "auxiliary_loss_clip": 0.0120243,
      "auxiliary_loss_mlp": 0.0103347,
      "balance_loss_clip": 0.98414892,
      "balance_loss_mlp": 1.02432084,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 2.647432441874539,
      "language_loss": 0.74689573,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.76925474,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 2.888288974761963
    },
    {
      "auxiliary_loss_clip": 0.0120999,
      "auxiliary_loss_mlp": 0.01036712,
      "balance_loss_clip": 1.02366197,
      "balance_loss_mlp": 1.02718067,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.0453315565993564,
      "language_loss": 0.69538456,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71785152,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.7051033973693848
    },
    {
      "auxiliary_loss_clip": 0.01198042,
      "auxiliary_loss_mlp": 0.01044641,
      "balance_loss_clip": 0.98072773,
      "balance_loss_mlp": 1.03535438,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 1.91132867138579,
      "language_loss": 0.83948433,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.86191112,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.6918416023254395
    },
    {
      "auxiliary_loss_clip": 0.01216061,
      "auxiliary_loss_mlp": 0.0103939,
      "balance_loss_clip": 0.98799431,
      "balance_loss_mlp": 1.02976334,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 1.7594849288946168,
      "language_loss": 0.73587084,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.7584253,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.7214295864105225
    },
    {
      "auxiliary_loss_clip": 0.01196988,
      "auxiliary_loss_mlp": 0.01126919,
      "balance_loss_clip": 0.9802897,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 1.8458286977216558,
      "language_loss": 0.67174858,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.69498765,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.6556503772735596
    },
    {
      "auxiliary_loss_clip": 0.01199636,
      "auxiliary_loss_mlp": 0.01036307,
      "balance_loss_clip": 0.82641232,
      "balance_loss_mlp": 1.02619219,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 1.709545456386331,
      "language_loss": 0.80704391,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.82940334,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 3.7950212955474854
    },
    {
      "auxiliary_loss_clip": 0.01212946,
      "auxiliary_loss_mlp": 0.01038937,
      "balance_loss_clip": 1.02526736,
      "balance_loss_mlp": 1.02935219,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 2.4441314714636144,
      "language_loss": 0.69019783,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.7127167,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 2.894369602203369
    },
    {
      "auxiliary_loss_clip": 0.0120542,
      "auxiliary_loss_mlp": 0.01034955,
      "balance_loss_clip": 1.02011919,
      "balance_loss_mlp": 1.02449489,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 1.5301933281604791,
      "language_loss": 0.73674059,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.75914431,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 3.5498745441436768
    },
    {
      "auxiliary_loss_clip": 0.01211696,
      "auxiliary_loss_mlp": 0.01031575,
      "balance_loss_clip": 1.0595727,
      "balance_loss_mlp": 1.02220476,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 2.0855764772466103,
      "language_loss": 0.77979892,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.80223167,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 2.6742167472839355
    },
    {
      "auxiliary_loss_clip": 0.01210522,
      "auxiliary_loss_mlp": 0.01039874,
      "balance_loss_clip": 1.06110525,
      "balance_loss_mlp": 1.02996826,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 1.807864446588758,
      "language_loss": 0.83288741,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85539126,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 3.459698438644409
    },
    {
      "auxiliary_loss_clip": 0.01202635,
      "auxiliary_loss_mlp": 0.01035653,
      "balance_loss_clip": 0.97899848,
      "balance_loss_mlp": 1.02658141,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 3.8722612385203683,
      "language_loss": 0.74761462,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.76999748,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 2.7302725315093994
    },
    {
      "auxiliary_loss_clip": 0.01212605,
      "auxiliary_loss_mlp": 0.01032599,
      "balance_loss_clip": 1.06047201,
      "balance_loss_mlp": 1.02355099,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 1.664052621643117,
      "language_loss": 0.80392861,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82638073,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 3.616485834121704
    },
    {
      "auxiliary_loss_clip": 0.01184536,
      "auxiliary_loss_mlp": 0.01035715,
      "balance_loss_clip": 0.86091495,
      "balance_loss_mlp": 1.02554595,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 2.5343676886405886,
      "language_loss": 0.67989904,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.70210153,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 3.1919617652893066
    },
    {
      "auxiliary_loss_clip": 0.01206733,
      "auxiliary_loss_mlp": 0.01037847,
      "balance_loss_clip": 1.02172399,
      "balance_loss_mlp": 1.02763116,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 1.9342419458132785,
      "language_loss": 0.83680356,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.85924935,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 3.2905945777893066
    },
    {
      "auxiliary_loss_clip": 0.01216763,
      "auxiliary_loss_mlp": 0.01126929,
      "balance_loss_clip": 0.947914,
      "balance_loss_mlp": 0.0,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.5060235980330963,
      "language_loss": 0.74460858,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.76804554,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.8780815601348877
    },
    {
      "auxiliary_loss_clip": 0.01199089,
      "auxiliary_loss_mlp": 0.01042254,
      "balance_loss_clip": 0.94193208,
      "balance_loss_mlp": 1.03234172,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 2.260897106185283,
      "language_loss": 0.81965435,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.84206772,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.7225847244262695
    },
    {
      "auxiliary_loss_clip": 0.0120621,
      "auxiliary_loss_mlp": 0.01039595,
      "balance_loss_clip": 0.94526237,
      "balance_loss_mlp": 1.02983212,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.2550224649061206,
      "language_loss": 0.80751193,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.82996994,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.893188714981079
    },
    {
      "auxiliary_loss_clip": 0.01204034,
      "auxiliary_loss_mlp": 0.01030239,
      "balance_loss_clip": 0.94288117,
      "balance_loss_mlp": 1.02078009,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 1.9301046222221194,
      "language_loss": 0.76155543,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.78389823,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.7424864768981934
    },
    {
      "auxiliary_loss_clip": 0.01182174,
      "auxiliary_loss_mlp": 0.01046404,
      "balance_loss_clip": 0.94123578,
      "balance_loss_mlp": 1.03713584,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 1.9575949611446293,
      "language_loss": 0.78416324,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.80644906,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.7290802001953125
    },
    {
      "auxiliary_loss_clip": 0.01196646,
      "auxiliary_loss_mlp": 0.01033333,
      "balance_loss_clip": 0.93950957,
      "balance_loss_mlp": 1.02371907,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 2.063123452147371,
      "language_loss": 0.69126201,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.71356183,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.689906358718872
    },
    {
      "auxiliary_loss_clip": 0.01202928,
      "auxiliary_loss_mlp": 0.01126102,
      "balance_loss_clip": 1.01812768,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 2.1091128885529433,
      "language_loss": 0.71170026,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.73499054,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.6940767765045166
    },
    {
      "auxiliary_loss_clip": 0.01218207,
      "auxiliary_loss_mlp": 0.01040386,
      "balance_loss_clip": 0.98225343,
      "balance_loss_mlp": 1.030128,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 2.387041902974419,
      "language_loss": 0.76883632,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.79142225,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.665356159210205
    },
    {
      "auxiliary_loss_clip": 0.01212017,
      "auxiliary_loss_mlp": 0.01032192,
      "balance_loss_clip": 1.02450919,
      "balance_loss_mlp": 1.02217829,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 3.5966978722091847,
      "language_loss": 0.72727752,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.74971962,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.7327752113342285
    },
    {
      "auxiliary_loss_clip": 0.01209266,
      "auxiliary_loss_mlp": 0.01034601,
      "balance_loss_clip": 1.01895666,
      "balance_loss_mlp": 1.0244863,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 3.4600309545828134,
      "language_loss": 0.75788581,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.78032446,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 2.9024386405944824
    },
    {
      "auxiliary_loss_clip": 0.01210249,
      "auxiliary_loss_mlp": 0.010359,
      "balance_loss_clip": 1.06020224,
      "balance_loss_mlp": 1.02620864,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 1.6497733362181726,
      "language_loss": 0.77961576,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.80207729,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 2.6504225730895996
    },
    {
      "auxiliary_loss_clip": 0.01208396,
      "auxiliary_loss_mlp": 0.01040018,
      "balance_loss_clip": 0.94371223,
      "balance_loss_mlp": 1.03029656,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 1.8320414077523788,
      "language_loss": 0.83397746,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85646164,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 2.7259299755096436
    },
    {
      "auxiliary_loss_clip": 0.01212304,
      "auxiliary_loss_mlp": 0.01032631,
      "balance_loss_clip": 1.0193727,
      "balance_loss_mlp": 1.02309418,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 1.9253695756529305,
      "language_loss": 0.81652701,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.83897638,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 2.6610617637634277
    },
    {
      "auxiliary_loss_clip": 0.0120915,
      "auxiliary_loss_mlp": 0.01040913,
      "balance_loss_clip": 1.06241417,
      "balance_loss_mlp": 1.03200758,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 2.9876520752072038,
      "language_loss": 0.82878947,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.85129011,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 2.678208827972412
    },
    {
      "auxiliary_loss_clip": 0.01189482,
      "auxiliary_loss_mlp": 0.01032936,
      "balance_loss_clip": 0.90065718,
      "balance_loss_mlp": 1.02345252,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 2.4110399123516144,
      "language_loss": 0.74108994,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76331413,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.831470489501953
    },
    {
      "auxiliary_loss_clip": 0.01201235,
      "auxiliary_loss_mlp": 0.01032215,
      "balance_loss_clip": 0.97720057,
      "balance_loss_mlp": 1.02257133,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 2.009472590317466,
      "language_loss": 0.78201365,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.80434817,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.841475248336792
    },
    {
      "auxiliary_loss_clip": 0.01205793,
      "auxiliary_loss_mlp": 0.01126632,
      "balance_loss_clip": 0.98494291,
      "balance_loss_mlp": 0.0,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 1.9718624752537304,
      "language_loss": 0.80336148,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.82668573,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.7041752338409424
    },
    {
      "auxiliary_loss_clip": 0.01210542,
      "auxiliary_loss_mlp": 0.01036036,
      "balance_loss_clip": 1.02065516,
      "balance_loss_mlp": 1.02646327,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 1.5471934796150149,
      "language_loss": 0.83101058,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.85347635,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 3.6215434074401855
    },
    {
      "auxiliary_loss_clip": 0.01207365,
      "auxiliary_loss_mlp": 0.01036205,
      "balance_loss_clip": 1.02115095,
      "balance_loss_mlp": 1.02591097,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 2.108820416585375,
      "language_loss": 0.72976434,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.75220007,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 2.7341110706329346
    },
    {
      "auxiliary_loss_clip": 0.01207067,
      "auxiliary_loss_mlp": 0.01038964,
      "balance_loss_clip": 0.98561656,
      "balance_loss_mlp": 1.02944541,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.0874358101376584,
      "language_loss": 0.69580078,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.71826106,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 3.676455497741699
    },
    {
      "auxiliary_loss_clip": 0.01209633,
      "auxiliary_loss_mlp": 0.0103325,
      "balance_loss_clip": 1.06020522,
      "balance_loss_mlp": 1.02400529,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 1.7614190266195446,
      "language_loss": 0.83575618,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.85818499,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 2.628675699234009
    },
    {
      "auxiliary_loss_clip": 0.01205116,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 0.98021972,
      "balance_loss_mlp": 1.0226922,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 2.0079350655219343,
      "language_loss": 0.83004379,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.85241461,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 3.6516616344451904
    },
    {
      "auxiliary_loss_clip": 0.01211356,
      "auxiliary_loss_mlp": 0.01040691,
      "balance_loss_clip": 1.060215,
      "balance_loss_mlp": 1.03062963,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 2.3665890014513025,
      "language_loss": 0.76128769,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78380823,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.6899218559265137
    },
    {
      "auxiliary_loss_clip": 0.01101207,
      "auxiliary_loss_mlp": 0.01003581,
      "balance_loss_clip": 1.02348113,
      "balance_loss_mlp": 0.9996714,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9017176317003461,
      "language_loss": 0.60573602,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62678397,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 3.8242404460906982
    },
    {
      "auxiliary_loss_clip": 0.01189196,
      "auxiliary_loss_mlp": 0.01030173,
      "balance_loss_clip": 0.94087559,
      "balance_loss_mlp": 1.0203557,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 1.8238278482777786,
      "language_loss": 0.74508095,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.76727468,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.776047468185425
    },
    {
      "auxiliary_loss_clip": 0.01206676,
      "auxiliary_loss_mlp": 0.01037089,
      "balance_loss_clip": 0.94334173,
      "balance_loss_mlp": 1.02740872,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 1.9110046736728326,
      "language_loss": 0.87741435,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.89985198,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 2.7478013038635254
    },
    {
      "auxiliary_loss_clip": 0.01197198,
      "auxiliary_loss_mlp": 0.01035589,
      "balance_loss_clip": 0.94387537,
      "balance_loss_mlp": 1.02600491,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 1.9257193787341087,
      "language_loss": 0.77454066,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79686856,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.7210426330566406
    },
    {
      "auxiliary_loss_clip": 0.01210202,
      "auxiliary_loss_mlp": 0.01030385,
      "balance_loss_clip": 1.02183962,
      "balance_loss_mlp": 1.02123606,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.5534309559802726,
      "language_loss": 0.8687582,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.89116406,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.790175676345825
    },
    {
      "auxiliary_loss_clip": 0.01211926,
      "auxiliary_loss_mlp": 0.01039742,
      "balance_loss_clip": 0.9444446,
      "balance_loss_mlp": 1.03016937,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 2.032751631432209,
      "language_loss": 0.88293588,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90545255,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.7470223903656006
    },
    {
      "auxiliary_loss_clip": 0.01200347,
      "auxiliary_loss_mlp": 0.01034429,
      "balance_loss_clip": 0.9398607,
      "balance_loss_mlp": 1.0246954,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 2.390223569136515,
      "language_loss": 0.61307245,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.6354202,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.7213878631591797
    },
    {
      "auxiliary_loss_clip": 0.01210055,
      "auxiliary_loss_mlp": 0.01033217,
      "balance_loss_clip": 0.98388362,
      "balance_loss_mlp": 1.02354908,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 2.1704261223026546,
      "language_loss": 0.81758368,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84001637,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.7118706703186035
    },
    {
      "auxiliary_loss_clip": 0.01194966,
      "auxiliary_loss_mlp": 0.01037361,
      "balance_loss_clip": 0.9799242,
      "balance_loss_mlp": 1.02752042,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 2.161799584818254,
      "language_loss": 0.73267877,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75500208,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.682037830352783
    },
    {
      "auxiliary_loss_clip": 0.01198306,
      "auxiliary_loss_mlp": 0.01036645,
      "balance_loss_clip": 0.98207045,
      "balance_loss_mlp": 1.02657723,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 1.8380057362141056,
      "language_loss": 0.85095716,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87330663,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.7195019721984863
    },
    {
      "auxiliary_loss_clip": 0.01198513,
      "auxiliary_loss_mlp": 0.0103233,
      "balance_loss_clip": 0.9846766,
      "balance_loss_mlp": 1.02309144,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.765049025404553,
      "language_loss": 0.81834817,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84065664,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.712526321411133
    },
    {
      "auxiliary_loss_clip": 0.0121332,
      "auxiliary_loss_mlp": 0.01035286,
      "balance_loss_clip": 1.06032205,
      "balance_loss_mlp": 1.02475953,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 2.5035011649946193,
      "language_loss": 0.77310604,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79559207,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.6190764904022217
    },
    {
      "auxiliary_loss_clip": 0.01207226,
      "auxiliary_loss_mlp": 0.0103311,
      "balance_loss_clip": 0.94133323,
      "balance_loss_mlp": 1.02378166,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 1.6695396068604949,
      "language_loss": 0.80168378,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.8240872,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.739814281463623
    },
    {
      "auxiliary_loss_clip": 0.01208773,
      "auxiliary_loss_mlp": 0.01032175,
      "balance_loss_clip": 0.94651937,
      "balance_loss_mlp": 1.02244806,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 2.9045469652640237,
      "language_loss": 0.73715109,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.75956059,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 2.7668871879577637
    },
    {
      "auxiliary_loss_clip": 0.0121205,
      "auxiliary_loss_mlp": 0.01034465,
      "balance_loss_clip": 1.05959177,
      "balance_loss_mlp": 1.0246954,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 1.8498612567937358,
      "language_loss": 0.75818735,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.78065252,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.652161121368408
    },
    {
      "auxiliary_loss_clip": 0.0120929,
      "auxiliary_loss_mlp": 0.01034386,
      "balance_loss_clip": 0.9862864,
      "balance_loss_mlp": 1.02510524,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 1.9812726278109136,
      "language_loss": 0.80611193,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.82854867,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 2.7372193336486816
    },
    {
      "auxiliary_loss_clip": 0.01201975,
      "auxiliary_loss_mlp": 0.010349,
      "balance_loss_clip": 0.98279947,
      "balance_loss_mlp": 1.02511883,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 2.730617581092196,
      "language_loss": 0.7080282,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.73039693,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 2.7442739009857178
    },
    {
      "auxiliary_loss_clip": 0.0120386,
      "auxiliary_loss_mlp": 0.01034235,
      "balance_loss_clip": 1.02215576,
      "balance_loss_mlp": 1.02419174,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 1.6360351459010283,
      "language_loss": 0.79333007,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81571102,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.719902515411377
    },
    {
      "auxiliary_loss_clip": 0.01206985,
      "auxiliary_loss_mlp": 0.01036271,
      "balance_loss_clip": 0.9442395,
      "balance_loss_mlp": 1.02650738,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 1.637844000825418,
      "language_loss": 0.88413072,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90656322,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.826120138168335
    },
    {
      "auxiliary_loss_clip": 0.0119416,
      "auxiliary_loss_mlp": 0.01040675,
      "balance_loss_clip": 0.94198471,
      "balance_loss_mlp": 1.03095901,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 1.9843717063649056,
      "language_loss": 0.83238012,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.8547284,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.7880430221557617
    },
    {
      "auxiliary_loss_clip": 0.01095965,
      "auxiliary_loss_mlp": 0.01121572,
      "balance_loss_clip": 0.94928157,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8734430420284602,
      "language_loss": 0.62691104,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64908648,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 4.250439405441284
    },
    {
      "auxiliary_loss_clip": 0.01211927,
      "auxiliary_loss_mlp": 0.01126816,
      "balance_loss_clip": 0.98558187,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 1.9102082932069804,
      "language_loss": 0.77489811,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.79828554,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 2.6951069831848145
    },
    {
      "auxiliary_loss_clip": 0.01214638,
      "auxiliary_loss_mlp": 0.0104239,
      "balance_loss_clip": 1.02341962,
      "balance_loss_mlp": 1.03214431,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 1.7767481806758199,
      "language_loss": 0.88756597,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.91013622,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 3.558506488800049
    },
    {
      "auxiliary_loss_clip": 0.01214872,
      "auxiliary_loss_mlp": 0.01042218,
      "balance_loss_clip": 1.06132746,
      "balance_loss_mlp": 1.03141797,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.6517042355042284,
      "language_loss": 0.69501901,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71758991,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 2.8427038192749023
    },
    {
      "auxiliary_loss_clip": 0.01099813,
      "auxiliary_loss_mlp": 0.01013384,
      "balance_loss_clip": 0.98626471,
      "balance_loss_mlp": 1.00956941,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7539604728525547,
      "language_loss": 0.54505104,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.56618303,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 4.011860132217407
    },
    {
      "auxiliary_loss_clip": 0.01205643,
      "auxiliary_loss_mlp": 0.01036127,
      "balance_loss_clip": 0.98391652,
      "balance_loss_mlp": 1.02599466,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 2.5520521690912634,
      "language_loss": 0.80891472,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.83133239,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.7138102054595947
    },
    {
      "auxiliary_loss_clip": 0.01200841,
      "auxiliary_loss_mlp": 0.01040322,
      "balance_loss_clip": 0.97930926,
      "balance_loss_mlp": 1.02921128,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 1.9257106380017777,
      "language_loss": 0.807024,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.82943559,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 3.6912097930908203
    },
    {
      "auxiliary_loss_clip": 0.01214735,
      "auxiliary_loss_mlp": 0.01035264,
      "balance_loss_clip": 1.06302023,
      "balance_loss_mlp": 1.0245831,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 2.2442679830330596,
      "language_loss": 0.76100957,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.78350949,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 2.6681714057922363
    },
    {
      "auxiliary_loss_clip": 0.01204815,
      "auxiliary_loss_mlp": 0.01038212,
      "balance_loss_clip": 0.90527171,
      "balance_loss_mlp": 1.02877092,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 1.695362424562407,
      "language_loss": 0.66459477,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.68702501,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.795278310775757
    },
    {
      "auxiliary_loss_clip": 0.01215152,
      "auxiliary_loss_mlp": 0.01032181,
      "balance_loss_clip": 1.02219987,
      "balance_loss_mlp": 1.0220542,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 4.909337009611394,
      "language_loss": 0.71478993,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.73726332,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.699272394180298
    },
    {
      "auxiliary_loss_clip": 0.01199855,
      "auxiliary_loss_mlp": 0.0104358,
      "balance_loss_clip": 0.94461203,
      "balance_loss_mlp": 1.03375673,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 1.8312330770176346,
      "language_loss": 0.78704959,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80948395,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.721095085144043
    },
    {
      "auxiliary_loss_clip": 0.01171708,
      "auxiliary_loss_mlp": 0.01040011,
      "balance_loss_clip": 0.86068308,
      "balance_loss_mlp": 1.02962744,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 1.7270512784528267,
      "language_loss": 0.79448992,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.81660718,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 3.028413772583008
    },
    {
      "auxiliary_loss_clip": 0.01211533,
      "auxiliary_loss_mlp": 0.01037179,
      "balance_loss_clip": 1.02215517,
      "balance_loss_mlp": 1.02769637,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 2.243605212233821,
      "language_loss": 0.8956629,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.91814995,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.99159836769104
    },
    {
      "auxiliary_loss_clip": 0.01194822,
      "auxiliary_loss_mlp": 0.01040296,
      "balance_loss_clip": 0.90683502,
      "balance_loss_mlp": 1.03015733,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 1.8012902433573421,
      "language_loss": 0.80851316,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.83086437,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.759506940841675
    },
    {
      "auxiliary_loss_clip": 0.01212824,
      "auxiliary_loss_mlp": 0.01034662,
      "balance_loss_clip": 1.06309664,
      "balance_loss_mlp": 1.02502966,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 1.6444946794599864,
      "language_loss": 0.66880715,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.69128203,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.707887887954712
    },
    {
      "auxiliary_loss_clip": 0.01202242,
      "auxiliary_loss_mlp": 0.01034687,
      "balance_loss_clip": 0.90224916,
      "balance_loss_mlp": 1.02422071,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 1.832851424372693,
      "language_loss": 0.7836591,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.80602837,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.8011701107025146
    },
    {
      "auxiliary_loss_clip": 0.01202911,
      "auxiliary_loss_mlp": 0.01032905,
      "balance_loss_clip": 0.98372138,
      "balance_loss_mlp": 1.0236187,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 2.9085276614546247,
      "language_loss": 0.66833788,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.690696,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.7325727939605713
    },
    {
      "auxiliary_loss_clip": 0.01209468,
      "auxiliary_loss_mlp": 0.01034223,
      "balance_loss_clip": 1.02409649,
      "balance_loss_mlp": 1.02448404,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.5198705294756183,
      "language_loss": 0.76738024,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.78981721,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.7005488872528076
    },
    {
      "auxiliary_loss_clip": 0.01112009,
      "auxiliary_loss_mlp": 0.01004839,
      "balance_loss_clip": 0.91865098,
      "balance_loss_mlp": 1.00121534,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.9108694660418466,
      "language_loss": 0.62711382,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.64828229,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.3971505165100098
    },
    {
      "auxiliary_loss_clip": 0.01208966,
      "auxiliary_loss_mlp": 0.01045479,
      "balance_loss_clip": 0.98540139,
      "balance_loss_mlp": 1.0352453,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 3.424137120617332,
      "language_loss": 1.00074339,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.02328777,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 2.6352477073669434
    },
    {
      "auxiliary_loss_clip": 0.01209761,
      "auxiliary_loss_mlp": 0.01026015,
      "balance_loss_clip": 1.02400637,
      "balance_loss_mlp": 1.0171454,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 1.6337176687420827,
      "language_loss": 0.78772771,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.81008542,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 2.6877546310424805
    },
    {
      "auxiliary_loss_clip": 0.01203827,
      "auxiliary_loss_mlp": 0.01036015,
      "balance_loss_clip": 0.9832778,
      "balance_loss_mlp": 1.02636456,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 2.1171998612923235,
      "language_loss": 0.8790462,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.90144461,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 2.6205854415893555
    },
    {
      "auxiliary_loss_clip": 0.01219448,
      "auxiliary_loss_mlp": 0.01037063,
      "balance_loss_clip": 0.98743653,
      "balance_loss_mlp": 1.02665019,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 4.467135859079109,
      "language_loss": 0.76168519,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78425038,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 2.7030646800994873
    },
    {
      "auxiliary_loss_clip": 0.01198254,
      "auxiliary_loss_mlp": 0.01034021,
      "balance_loss_clip": 0.98412597,
      "balance_loss_mlp": 1.0242399,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 2.2467599381869774,
      "language_loss": 0.91035581,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.93267858,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.6511154174804688
    },
    {
      "auxiliary_loss_clip": 0.01217172,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.06661236,
      "balance_loss_mlp": 1.02416945,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 1.7798122890926995,
      "language_loss": 0.88500988,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.90751827,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 2.685845375061035
    },
    {
      "auxiliary_loss_clip": 0.01207943,
      "auxiliary_loss_mlp": 0.01031567,
      "balance_loss_clip": 0.98718083,
      "balance_loss_mlp": 1.02219701,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.6197000294546855,
      "language_loss": 0.77587748,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.79827261,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.783418655395508
    },
    {
      "auxiliary_loss_clip": 0.01103709,
      "auxiliary_loss_mlp": 0.01002541,
      "balance_loss_clip": 0.9111973,
      "balance_loss_mlp": 0.99884558,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6673111375177618,
      "language_loss": 0.56931531,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.59037781,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 4.280874252319336
    },
    {
      "auxiliary_loss_clip": 0.01192417,
      "auxiliary_loss_mlp": 0.01040039,
      "balance_loss_clip": 0.94111246,
      "balance_loss_mlp": 1.03099072,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 1.9093458457080674,
      "language_loss": 0.83742982,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85975432,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 2.780397415161133
    },
    {
      "auxiliary_loss_clip": 0.01201325,
      "auxiliary_loss_mlp": 0.01036816,
      "balance_loss_clip": 1.02359414,
      "balance_loss_mlp": 1.02665305,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 1.515350390709042,
      "language_loss": 0.81221133,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.8345927,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 3.666937828063965
    },
    {
      "auxiliary_loss_clip": 0.01208967,
      "auxiliary_loss_mlp": 0.01034888,
      "balance_loss_clip": 0.98704261,
      "balance_loss_mlp": 1.02483284,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 2.8130215005647043,
      "language_loss": 0.86471415,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.88715267,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 2.617405414581299
    },
    {
      "auxiliary_loss_clip": 0.0120237,
      "auxiliary_loss_mlp": 0.01038501,
      "balance_loss_clip": 0.86838663,
      "balance_loss_mlp": 1.02886295,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 1.827809315924596,
      "language_loss": 0.71323097,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73563975,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.9949851036071777
    },
    {
      "auxiliary_loss_clip": 0.01212509,
      "auxiliary_loss_mlp": 0.01031322,
      "balance_loss_clip": 0.91019249,
      "balance_loss_mlp": 1.02191615,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.7783105582569263,
      "language_loss": 0.82025611,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.8426944,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 2.8097712993621826
    },
    {
      "auxiliary_loss_clip": 0.01104007,
      "auxiliary_loss_mlp": 0.01006502,
      "balance_loss_clip": 0.8744539,
      "balance_loss_mlp": 1.00304461,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9178917123252953,
      "language_loss": 0.56643653,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58754158,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 4.236000061035156
    },
    {
      "auxiliary_loss_clip": 0.01216135,
      "auxiliary_loss_mlp": 0.01038293,
      "balance_loss_clip": 1.02496397,
      "balance_loss_mlp": 1.02820802,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 3.1689167027685894,
      "language_loss": 0.70908463,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.73162889,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.602689266204834
    },
    {
      "auxiliary_loss_clip": 0.01199348,
      "auxiliary_loss_mlp": 0.01035719,
      "balance_loss_clip": 0.98066008,
      "balance_loss_mlp": 1.02544868,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.4221179066058776,
      "language_loss": 0.67222631,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69457686,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 2.6903045177459717
    },
    {
      "auxiliary_loss_clip": 0.01199373,
      "auxiliary_loss_mlp": 0.01038537,
      "balance_loss_clip": 0.98122638,
      "balance_loss_mlp": 1.02859449,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 1.7832567099666734,
      "language_loss": 0.80069089,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.82306999,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.6555209159851074
    },
    {
      "auxiliary_loss_clip": 0.01209801,
      "auxiliary_loss_mlp": 0.01041289,
      "balance_loss_clip": 1.02229869,
      "balance_loss_mlp": 1.03129935,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 1.6116927813786062,
      "language_loss": 0.75534934,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.77786028,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.6999635696411133
    },
    {
      "auxiliary_loss_clip": 0.01203989,
      "auxiliary_loss_mlp": 0.01127114,
      "balance_loss_clip": 0.98097205,
      "balance_loss_mlp": 0.0,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 1.8430082220818174,
      "language_loss": 0.82363451,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.84694558,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.7454376220703125
    },
    {
      "auxiliary_loss_clip": 0.01209688,
      "auxiliary_loss_mlp": 0.01032593,
      "balance_loss_clip": 0.98240858,
      "balance_loss_mlp": 1.02266312,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 1.8931240000231084,
      "language_loss": 0.89480942,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.91723216,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.721705436706543
    },
    {
      "auxiliary_loss_clip": 0.01207878,
      "auxiliary_loss_mlp": 0.01031198,
      "balance_loss_clip": 0.98620605,
      "balance_loss_mlp": 1.02231073,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 1.797769581686244,
      "language_loss": 0.75817668,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.78056741,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.781683921813965
    },
    {
      "auxiliary_loss_clip": 0.01213843,
      "auxiliary_loss_mlp": 0.01038613,
      "balance_loss_clip": 1.02317977,
      "balance_loss_mlp": 1.02843881,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 2.0311212443997606,
      "language_loss": 0.84771615,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.87024069,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.7412893772125244
    },
    {
      "auxiliary_loss_clip": 0.01215853,
      "auxiliary_loss_mlp": 0.01040412,
      "balance_loss_clip": 1.02340865,
      "balance_loss_mlp": 1.03065467,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 1.9089684099040418,
      "language_loss": 0.84300214,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86556476,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.6874618530273438
    },
    {
      "auxiliary_loss_clip": 0.01196145,
      "auxiliary_loss_mlp": 0.01036545,
      "balance_loss_clip": 0.93971503,
      "balance_loss_mlp": 1.0263164,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 1.84611136329881,
      "language_loss": 0.78434289,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.80666983,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.793020486831665
    },
    {
      "auxiliary_loss_clip": 0.01209856,
      "auxiliary_loss_mlp": 0.01033234,
      "balance_loss_clip": 1.06029344,
      "balance_loss_mlp": 1.02363157,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 1.574590612994341,
      "language_loss": 0.81158555,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.83401644,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.6638941764831543
    },
    {
      "auxiliary_loss_clip": 0.01199804,
      "auxiliary_loss_mlp": 0.01034119,
      "balance_loss_clip": 0.86998808,
      "balance_loss_mlp": 1.02416468,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 2.377708821575719,
      "language_loss": 0.76914573,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.79148495,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.796387195587158
    },
    {
      "auxiliary_loss_clip": 0.01211898,
      "auxiliary_loss_mlp": 0.01037319,
      "balance_loss_clip": 1.06027687,
      "balance_loss_mlp": 1.02784252,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 1.9200869364221267,
      "language_loss": 0.83329749,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.85578966,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.655673027038574
    },
    {
      "auxiliary_loss_clip": 0.01207775,
      "auxiliary_loss_mlp": 0.01040065,
      "balance_loss_clip": 0.98395669,
      "balance_loss_mlp": 1.02950883,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 1.9177399074078099,
      "language_loss": 0.79883766,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82131606,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.687455415725708
    },
    {
      "auxiliary_loss_clip": 0.01197333,
      "auxiliary_loss_mlp": 0.01034059,
      "balance_loss_clip": 1.02142191,
      "balance_loss_mlp": 1.02414072,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 1.8648758258500078,
      "language_loss": 0.9004311,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.92274505,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 2.7164738178253174
    },
    {
      "auxiliary_loss_clip": 0.01197199,
      "auxiliary_loss_mlp": 0.01038176,
      "balance_loss_clip": 0.98251057,
      "balance_loss_mlp": 1.02785277,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 1.5912999678261013,
      "language_loss": 0.79123819,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81359196,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 2.650212049484253
    },
    {
      "auxiliary_loss_clip": 0.0120481,
      "auxiliary_loss_mlp": 0.01034795,
      "balance_loss_clip": 0.98478711,
      "balance_loss_mlp": 1.02477002,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 2.009818792991254,
      "language_loss": 0.80628753,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.82868361,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 2.6738054752349854
    },
    {
      "auxiliary_loss_clip": 0.01095947,
      "auxiliary_loss_mlp": 0.01008201,
      "balance_loss_clip": 0.98481417,
      "balance_loss_mlp": 1.00491083,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9709112143176121,
      "language_loss": 0.58157903,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60262054,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.2452547550201416
    },
    {
      "auxiliary_loss_clip": 0.01214781,
      "auxiliary_loss_mlp": 0.01041638,
      "balance_loss_clip": 1.0255971,
      "balance_loss_mlp": 1.03173721,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 1.785391781796221,
      "language_loss": 0.70935929,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.73192346,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.669532299041748
    },
    {
      "auxiliary_loss_clip": 0.01197615,
      "auxiliary_loss_mlp": 0.01041968,
      "balance_loss_clip": 0.98139608,
      "balance_loss_mlp": 1.03164411,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 1.6834884086122888,
      "language_loss": 0.8053292,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82772505,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 3.750321865081787
    },
    {
      "auxiliary_loss_clip": 0.01201881,
      "auxiliary_loss_mlp": 0.01036157,
      "balance_loss_clip": 0.94392288,
      "balance_loss_mlp": 1.02576208,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 1.7531246221631684,
      "language_loss": 0.70659077,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.72897112,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.7849793434143066
    },
    {
      "auxiliary_loss_clip": 0.01102728,
      "auxiliary_loss_mlp": 0.01009869,
      "balance_loss_clip": 0.98672056,
      "balance_loss_mlp": 1.00619721,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.9037392768892412,
      "language_loss": 0.63587344,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65699935,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 4.103297472000122
    },
    {
      "auxiliary_loss_clip": 0.01100257,
      "auxiliary_loss_mlp": 0.01004001,
      "balance_loss_clip": 1.02435541,
      "balance_loss_mlp": 1.00028121,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8434043630258714,
      "language_loss": 0.55826384,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.57930636,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 3.275162696838379
    },
    {
      "auxiliary_loss_clip": 0.01211727,
      "auxiliary_loss_mlp": 0.01034716,
      "balance_loss_clip": 1.02332139,
      "balance_loss_mlp": 1.02536392,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 2.1285991359726713,
      "language_loss": 0.74518263,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.76764703,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.7138874530792236
    },
    {
      "auxiliary_loss_clip": 0.01205219,
      "auxiliary_loss_mlp": 0.01034717,
      "balance_loss_clip": 0.9812541,
      "balance_loss_mlp": 1.02541804,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 2.273241685222256,
      "language_loss": 0.84092647,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86332583,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 3.799924373626709
    },
    {
      "auxiliary_loss_clip": 0.01210053,
      "auxiliary_loss_mlp": 0.01034467,
      "balance_loss_clip": 1.05886149,
      "balance_loss_mlp": 1.0242393,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 2.0289648812122336,
      "language_loss": 0.79385394,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81629914,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 3.6121034622192383
    },
    {
      "auxiliary_loss_clip": 0.01193586,
      "auxiliary_loss_mlp": 0.01039176,
      "balance_loss_clip": 0.90161836,
      "balance_loss_mlp": 1.02957964,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 1.6234669917246332,
      "language_loss": 0.7109701,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.73329771,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.740567922592163
    },
    {
      "auxiliary_loss_clip": 0.01210431,
      "auxiliary_loss_mlp": 0.01040083,
      "balance_loss_clip": 1.06060719,
      "balance_loss_mlp": 1.03005767,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 2.8035771796143196,
      "language_loss": 0.80171061,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.82421571,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 2.5949506759643555
    },
    {
      "auxiliary_loss_clip": 0.01210716,
      "auxiliary_loss_mlp": 0.01042246,
      "balance_loss_clip": 0.98455667,
      "balance_loss_mlp": 1.03251898,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 3.4314641915379296,
      "language_loss": 0.81745088,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.83998048,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 2.686755418777466
    },
    {
      "auxiliary_loss_clip": 0.0121416,
      "auxiliary_loss_mlp": 0.01036883,
      "balance_loss_clip": 1.02002597,
      "balance_loss_mlp": 1.02654767,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 32.182704003772585,
      "language_loss": 0.76625365,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78876412,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.628180503845215
    },
    {
      "auxiliary_loss_clip": 0.01206381,
      "auxiliary_loss_mlp": 0.01038091,
      "balance_loss_clip": 0.94397807,
      "balance_loss_mlp": 1.02797627,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 2.1612118824078097,
      "language_loss": 0.69000483,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.71244955,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 2.6950342655181885
    },
    {
      "auxiliary_loss_clip": 0.01207649,
      "auxiliary_loss_mlp": 0.01031377,
      "balance_loss_clip": 1.02430809,
      "balance_loss_mlp": 1.02226329,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 2.3663652599963982,
      "language_loss": 0.82424009,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84663033,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.6825640201568604
    },
    {
      "auxiliary_loss_clip": 0.01205022,
      "auxiliary_loss_mlp": 0.01127193,
      "balance_loss_clip": 0.98259962,
      "balance_loss_mlp": 0.0,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 1.8230191390974537,
      "language_loss": 0.78854042,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.81186259,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.7093405723571777
    },
    {
      "auxiliary_loss_clip": 0.01204664,
      "auxiliary_loss_mlp": 0.01031071,
      "balance_loss_clip": 0.98327076,
      "balance_loss_mlp": 1.0212543,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 1.9894769097588203,
      "language_loss": 0.79958689,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.82194424,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.69791316986084
    },
    {
      "auxiliary_loss_clip": 0.01094263,
      "auxiliary_loss_mlp": 0.0100314,
      "balance_loss_clip": 0.86840439,
      "balance_loss_mlp": 0.99946862,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.8662888932895882,
      "language_loss": 0.5778814,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59885544,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.414092540740967
    },
    {
      "auxiliary_loss_clip": 0.01206125,
      "auxiliary_loss_mlp": 0.01034409,
      "balance_loss_clip": 1.02250433,
      "balance_loss_mlp": 1.02575445,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 1.4541023805293876,
      "language_loss": 0.86565697,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.88806236,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.9329166412353516
    },
    {
      "auxiliary_loss_clip": 0.01208801,
      "auxiliary_loss_mlp": 0.0103422,
      "balance_loss_clip": 1.02071857,
      "balance_loss_mlp": 1.02482677,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.655405567480732,
      "language_loss": 0.85229015,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.87472034,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.8022208213806152
    },
    {
      "auxiliary_loss_clip": 0.01199432,
      "auxiliary_loss_mlp": 0.0103718,
      "balance_loss_clip": 0.94347286,
      "balance_loss_mlp": 1.02689254,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.0741035043804756,
      "language_loss": 0.7843498,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.80671591,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.7085072994232178
    },
    {
      "auxiliary_loss_clip": 0.01210454,
      "auxiliary_loss_mlp": 0.0103153,
      "balance_loss_clip": 1.06029963,
      "balance_loss_mlp": 1.02257705,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 1.7384218555109279,
      "language_loss": 0.80132604,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.82374585,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.761719226837158
    },
    {
      "auxiliary_loss_clip": 0.0121449,
      "auxiliary_loss_mlp": 0.01042891,
      "balance_loss_clip": 1.06526864,
      "balance_loss_mlp": 1.03367603,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 1.525982118035156,
      "language_loss": 0.78377229,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.80634606,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.6456234455108643
    },
    {
      "auxiliary_loss_clip": 0.01205835,
      "auxiliary_loss_mlp": 0.01033333,
      "balance_loss_clip": 0.98617673,
      "balance_loss_mlp": 1.02418375,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 1.674183890816899,
      "language_loss": 0.68803763,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71042931,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 2.9638280868530273
    },
    {
      "auxiliary_loss_clip": 0.01198333,
      "auxiliary_loss_mlp": 0.01028544,
      "balance_loss_clip": 0.86464816,
      "balance_loss_mlp": 1.01812482,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 1.8093237270375608,
      "language_loss": 0.77213383,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.7944026,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 2.8049209117889404
    },
    {
      "auxiliary_loss_clip": 0.01207034,
      "auxiliary_loss_mlp": 0.010317,
      "balance_loss_clip": 1.02406144,
      "balance_loss_mlp": 1.02203822,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.8228196706823523,
      "language_loss": 0.85301781,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87540519,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 2.6705691814422607
    },
    {
      "auxiliary_loss_clip": 0.01104842,
      "auxiliary_loss_mlp": 0.01002519,
      "balance_loss_clip": 0.94835806,
      "balance_loss_mlp": 0.9990142,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 1.0076723071405032,
      "language_loss": 0.63222468,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65329832,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.247112512588501
    },
    {
      "auxiliary_loss_clip": 0.01196652,
      "auxiliary_loss_mlp": 0.01040841,
      "balance_loss_clip": 0.9826771,
      "balance_loss_mlp": 1.03094029,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 1.8899357901707419,
      "language_loss": 0.77553189,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.79790688,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.7063212394714355
    },
    {
      "auxiliary_loss_clip": 0.01205259,
      "auxiliary_loss_mlp": 0.01035703,
      "balance_loss_clip": 0.98270011,
      "balance_loss_mlp": 1.02604723,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 2.0457211279844554,
      "language_loss": 0.83728182,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.85969144,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 4.518396377563477
    },
    {
      "auxiliary_loss_clip": 0.0120561,
      "auxiliary_loss_mlp": 0.01036194,
      "balance_loss_clip": 0.97731465,
      "balance_loss_mlp": 1.0258348,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 1.7562201095030658,
      "language_loss": 0.81242061,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.83483863,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.8744771480560303
    },
    {
      "auxiliary_loss_clip": 0.01208229,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 0.94446719,
      "balance_loss_mlp": 1.02014995,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 1.8356991342918316,
      "language_loss": 0.86499107,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.88736904,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 3.693739175796509
    },
    {
      "auxiliary_loss_clip": 0.01203514,
      "auxiliary_loss_mlp": 0.01035279,
      "balance_loss_clip": 0.9443298,
      "balance_loss_mlp": 1.02603483,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 1.627664415120511,
      "language_loss": 0.91126037,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93364829,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 2.864478588104248
    },
    {
      "auxiliary_loss_clip": 0.0120298,
      "auxiliary_loss_mlp": 0.01034437,
      "balance_loss_clip": 0.98584759,
      "balance_loss_mlp": 1.02497816,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 1.8867658998148815,
      "language_loss": 0.77729213,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.79966629,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 3.789491653442383
    },
    {
      "auxiliary_loss_clip": 0.01212787,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.06434953,
      "balance_loss_mlp": 1.02371275,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.5952438753624787,
      "language_loss": 0.77972555,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80218142,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 2.7274551391601562
    },
    {
      "auxiliary_loss_clip": 0.01199818,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.02191663,
      "balance_loss_mlp": 1.02158844,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 2.362855930644652,
      "language_loss": 0.73069489,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75300848,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 3.6694419384002686
    },
    {
      "auxiliary_loss_clip": 0.0119191,
      "auxiliary_loss_mlp": 0.01030407,
      "balance_loss_clip": 0.98131883,
      "balance_loss_mlp": 1.02086473,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 1.6630361269012979,
      "language_loss": 0.59808576,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.62030894,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.961310863494873
    },
    {
      "auxiliary_loss_clip": 0.01205998,
      "auxiliary_loss_mlp": 0.01036629,
      "balance_loss_clip": 0.98485976,
      "balance_loss_mlp": 1.02737284,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 2.301086867791505,
      "language_loss": 0.7571612,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.77958751,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.7277894020080566
    },
    {
      "auxiliary_loss_clip": 0.01206435,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 0.94529963,
      "balance_loss_mlp": 1.02116024,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 2.089187233761193,
      "language_loss": 0.87449783,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.89686799,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 2.822000741958618
    },
    {
      "auxiliary_loss_clip": 0.01190498,
      "auxiliary_loss_mlp": 0.01126654,
      "balance_loss_clip": 0.86064869,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 15.473837650117128,
      "language_loss": 0.74737364,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.77054513,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.863391399383545
    },
    {
      "auxiliary_loss_clip": 0.01190098,
      "auxiliary_loss_mlp": 0.01127008,
      "balance_loss_clip": 0.90246534,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 2.040816439340186,
      "language_loss": 0.74320859,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.76637959,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.8568179607391357
    },
    {
      "auxiliary_loss_clip": 0.01212297,
      "auxiliary_loss_mlp": 0.01032529,
      "balance_loss_clip": 0.98379493,
      "balance_loss_mlp": 1.0225811,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 1.9054959910213913,
      "language_loss": 0.8334983,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85594654,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.7329509258270264
    },
    {
      "auxiliary_loss_clip": 0.0121004,
      "auxiliary_loss_mlp": 0.01126832,
      "balance_loss_clip": 1.02101564,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 2.6919157479236784,
      "language_loss": 0.80497003,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.82833868,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.804258108139038
    },
    {
      "auxiliary_loss_clip": 0.01201504,
      "auxiliary_loss_mlp": 0.01038172,
      "balance_loss_clip": 0.98328018,
      "balance_loss_mlp": 1.02812839,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 1.8084565761319522,
      "language_loss": 0.84317034,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86556709,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.765552520751953
    },
    {
      "auxiliary_loss_clip": 0.01101393,
      "auxiliary_loss_mlp": 0.01018219,
      "balance_loss_clip": 1.02643228,
      "balance_loss_mlp": 1.01469028,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.8468182209610944,
      "language_loss": 0.52726865,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54846478,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.2334909439086914
    },
    {
      "auxiliary_loss_clip": 0.01200536,
      "auxiliary_loss_mlp": 0.01034322,
      "balance_loss_clip": 0.94439697,
      "balance_loss_mlp": 1.02473748,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 1.6166670679131998,
      "language_loss": 0.77304566,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.79539418,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.8569414615631104
    },
    {
      "auxiliary_loss_clip": 0.01194127,
      "auxiliary_loss_mlp": 0.01038419,
      "balance_loss_clip": 0.78849757,
      "balance_loss_mlp": 1.02785683,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 2.2918407374836205,
      "language_loss": 0.75420159,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77652705,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 3.1000351905822754
    },
    {
      "auxiliary_loss_clip": 0.01209214,
      "auxiliary_loss_mlp": 0.01041207,
      "balance_loss_clip": 1.0247848,
      "balance_loss_mlp": 1.03156257,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 1.7236040416427687,
      "language_loss": 0.79318106,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.81568527,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.999743700027466
    },
    {
      "auxiliary_loss_clip": 0.01213397,
      "auxiliary_loss_mlp": 0.01034553,
      "balance_loss_clip": 0.98505008,
      "balance_loss_mlp": 1.02377009,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 1.9750962836176396,
      "language_loss": 0.81755912,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.8400386,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.8750534057617188
    },
    {
      "auxiliary_loss_clip": 0.01213855,
      "auxiliary_loss_mlp": 0.01036665,
      "balance_loss_clip": 0.98521519,
      "balance_loss_mlp": 1.02567387,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 2.8481992146184654,
      "language_loss": 0.80722064,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.82972586,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 2.8706490993499756
    },
    {
      "auxiliary_loss_clip": 0.0121222,
      "auxiliary_loss_mlp": 0.011264,
      "balance_loss_clip": 1.06222606,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 1.6860951399749857,
      "language_loss": 0.73822689,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.76161301,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 2.747596025466919
    },
    {
      "auxiliary_loss_clip": 0.01105743,
      "auxiliary_loss_mlp": 0.01009968,
      "balance_loss_clip": 0.98963523,
      "balance_loss_mlp": 1.00651073,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7891079115406868,
      "language_loss": 0.64850724,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66966432,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 3.140472650527954
    },
    {
      "auxiliary_loss_clip": 0.01196764,
      "auxiliary_loss_mlp": 0.0103347,
      "balance_loss_clip": 0.98103833,
      "balance_loss_mlp": 1.02378452,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 1.78649631640306,
      "language_loss": 0.89676642,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.91906881,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 2.7592086791992188
    },
    {
      "auxiliary_loss_clip": 0.0121112,
      "auxiliary_loss_mlp": 0.0103352,
      "balance_loss_clip": 1.06278563,
      "balance_loss_mlp": 1.02323794,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 1.71380693791751,
      "language_loss": 0.8119505,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.8343969,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.727023124694824
    },
    {
      "auxiliary_loss_clip": 0.01214374,
      "auxiliary_loss_mlp": 0.01039147,
      "balance_loss_clip": 1.06178737,
      "balance_loss_mlp": 1.02876949,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 2.664054966503023,
      "language_loss": 0.71594667,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.73848188,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.652543783187866
    },
    {
      "auxiliary_loss_clip": 0.01205389,
      "auxiliary_loss_mlp": 0.01032368,
      "balance_loss_clip": 0.98616374,
      "balance_loss_mlp": 1.02272451,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 1.9162753157081192,
      "language_loss": 0.78429484,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.80667239,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 3.960226535797119
    },
    {
      "auxiliary_loss_clip": 0.01206737,
      "auxiliary_loss_mlp": 0.01040449,
      "balance_loss_clip": 0.90414059,
      "balance_loss_mlp": 1.03004253,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 1.8214104559277116,
      "language_loss": 0.83236885,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.85484076,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.846723794937134
    },
    {
      "auxiliary_loss_clip": 0.01209074,
      "auxiliary_loss_mlp": 0.01034482,
      "balance_loss_clip": 0.98532867,
      "balance_loss_mlp": 1.02412868,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.050344708547424,
      "language_loss": 0.82083189,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84326744,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 2.789783000946045
    },
    {
      "auxiliary_loss_clip": 0.01210106,
      "auxiliary_loss_mlp": 0.01041029,
      "balance_loss_clip": 1.024611,
      "balance_loss_mlp": 1.03143311,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 1.815458480799987,
      "language_loss": 0.88316351,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90567487,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 3.743311643600464
    },
    {
      "auxiliary_loss_clip": 0.01189435,
      "auxiliary_loss_mlp": 0.01029881,
      "balance_loss_clip": 0.7906059,
      "balance_loss_mlp": 1.02015924,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 1.6308478500909276,
      "language_loss": 0.77118421,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79337734,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 3.063040256500244
    },
    {
      "auxiliary_loss_clip": 0.0120809,
      "auxiliary_loss_mlp": 0.01036462,
      "balance_loss_clip": 1.02399397,
      "balance_loss_mlp": 1.02724755,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.3080017323999935,
      "language_loss": 0.75365591,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.77610147,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 3.973482847213745
    },
    {
      "auxiliary_loss_clip": 0.01195016,
      "auxiliary_loss_mlp": 0.01041473,
      "balance_loss_clip": 0.90711886,
      "balance_loss_mlp": 1.03166771,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 1.7638625669471528,
      "language_loss": 0.70664322,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.72900814,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 2.904266357421875
    },
    {
      "auxiliary_loss_clip": 0.01190631,
      "auxiliary_loss_mlp": 0.01039088,
      "balance_loss_clip": 0.90106797,
      "balance_loss_mlp": 1.02897942,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 3.8151673673898854,
      "language_loss": 0.74417657,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76647377,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 3.7527801990509033
    },
    {
      "auxiliary_loss_clip": 0.01209085,
      "auxiliary_loss_mlp": 0.01038929,
      "balance_loss_clip": 0.94408834,
      "balance_loss_mlp": 1.02928483,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.5179487156740175,
      "language_loss": 0.81840873,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.84088892,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.962275981903076
    },
    {
      "auxiliary_loss_clip": 0.01093953,
      "auxiliary_loss_mlp": 0.01003282,
      "balance_loss_clip": 0.8338446,
      "balance_loss_mlp": 0.99992001,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9110523772065345,
      "language_loss": 0.60941386,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.63038623,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 3.478878974914551
    },
    {
      "auxiliary_loss_clip": 0.01204804,
      "auxiliary_loss_mlp": 0.01037201,
      "balance_loss_clip": 1.02342129,
      "balance_loss_mlp": 1.02794456,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 1.7493382931789319,
      "language_loss": 0.83666581,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.85908586,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 2.859805107116699
    },
    {
      "auxiliary_loss_clip": 0.01205357,
      "auxiliary_loss_mlp": 0.0103248,
      "balance_loss_clip": 0.98433453,
      "balance_loss_mlp": 1.02277601,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 1.9175918674500723,
      "language_loss": 0.74736977,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.76974815,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.7947919368743896
    },
    {
      "auxiliary_loss_clip": 0.01212098,
      "auxiliary_loss_mlp": 0.01033047,
      "balance_loss_clip": 0.90708762,
      "balance_loss_mlp": 1.02312231,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 1.8967976062378902,
      "language_loss": 0.83224642,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.85469782,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 2.8184633255004883
    },
    {
      "auxiliary_loss_clip": 0.01210387,
      "auxiliary_loss_mlp": 0.01036751,
      "balance_loss_clip": 1.06056654,
      "balance_loss_mlp": 1.02677894,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 2.0991371360245235,
      "language_loss": 0.869317,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.89178842,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.7064013481140137
    },
    {
      "auxiliary_loss_clip": 0.01206076,
      "auxiliary_loss_mlp": 0.01034276,
      "balance_loss_clip": 0.98465931,
      "balance_loss_mlp": 1.02507257,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.5188989725225912,
      "language_loss": 0.79879951,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.82120305,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.742399215698242
    },
    {
      "auxiliary_loss_clip": 0.01208459,
      "auxiliary_loss_mlp": 0.01036598,
      "balance_loss_clip": 1.02266121,
      "balance_loss_mlp": 1.02737069,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 1.7890118617440696,
      "language_loss": 0.75239456,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77484506,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.7729227542877197
    },
    {
      "auxiliary_loss_clip": 0.01208331,
      "auxiliary_loss_mlp": 0.01034464,
      "balance_loss_clip": 1.06174862,
      "balance_loss_mlp": 1.0252012,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.4253821291792215,
      "language_loss": 0.84028995,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.86271787,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.655189037322998
    },
    {
      "auxiliary_loss_clip": 0.01210567,
      "auxiliary_loss_mlp": 0.01047515,
      "balance_loss_clip": 1.02514136,
      "balance_loss_mlp": 1.03788304,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 1.8656633146000465,
      "language_loss": 0.89289939,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.91548026,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.763049602508545
    },
    {
      "auxiliary_loss_clip": 0.01210712,
      "auxiliary_loss_mlp": 0.01039157,
      "balance_loss_clip": 1.06435418,
      "balance_loss_mlp": 1.02976322,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 2.0885827912964694,
      "language_loss": 0.7351293,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75762796,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.747924327850342
    },
    {
      "auxiliary_loss_clip": 0.01210982,
      "auxiliary_loss_mlp": 0.01035122,
      "balance_loss_clip": 0.95137858,
      "balance_loss_mlp": 1.02542448,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 2.5636393297952087,
      "language_loss": 0.92310011,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94556111,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.8180859088897705
    },
    {
      "auxiliary_loss_clip": 0.01206505,
      "auxiliary_loss_mlp": 0.01030289,
      "balance_loss_clip": 0.98756021,
      "balance_loss_mlp": 1.02087188,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.6110725874528256,
      "language_loss": 0.82210648,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.84447443,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 2.7517142295837402
    },
    {
      "auxiliary_loss_clip": 0.01206416,
      "auxiliary_loss_mlp": 0.01042288,
      "balance_loss_clip": 1.02372134,
      "balance_loss_mlp": 1.03331208,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 1.9975820311764305,
      "language_loss": 0.87182671,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89431375,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 2.6823909282684326
    },
    {
      "auxiliary_loss_clip": 0.01206272,
      "auxiliary_loss_mlp": 0.01036606,
      "balance_loss_clip": 0.94724882,
      "balance_loss_mlp": 1.02633619,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.1706944149252463,
      "language_loss": 0.82700455,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.84943336,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 2.8131089210510254
    },
    {
      "auxiliary_loss_clip": 0.01204852,
      "auxiliary_loss_mlp": 0.0103526,
      "balance_loss_clip": 0.90276319,
      "balance_loss_mlp": 1.02543116,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.6648014406015428,
      "language_loss": 0.78301007,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80541116,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 2.7739295959472656
    },
    {
      "auxiliary_loss_clip": 0.01193517,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 0.86699247,
      "balance_loss_mlp": 1.02201104,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 1.7530765979186083,
      "language_loss": 0.86393869,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88619375,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.966953754425049
    },
    {
      "auxiliary_loss_clip": 0.01208458,
      "auxiliary_loss_mlp": 0.01030638,
      "balance_loss_clip": 1.06132698,
      "balance_loss_mlp": 1.02146494,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 2.1924567465546274,
      "language_loss": 0.82460696,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84699786,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 3.7514569759368896
    },
    {
      "auxiliary_loss_clip": 0.0110507,
      "auxiliary_loss_mlp": 0.01004029,
      "balance_loss_clip": 0.90974438,
      "balance_loss_mlp": 1.00071478,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.806645193581211,
      "language_loss": 0.60350442,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.6245954,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.268483877182007
    },
    {
      "auxiliary_loss_clip": 0.01203467,
      "auxiliary_loss_mlp": 0.01033282,
      "balance_loss_clip": 1.02108383,
      "balance_loss_mlp": 1.02356637,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 1.9449035894786153,
      "language_loss": 0.82822764,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.85059512,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.7599217891693115
    },
    {
      "auxiliary_loss_clip": 0.01202653,
      "auxiliary_loss_mlp": 0.01032769,
      "balance_loss_clip": 0.98390663,
      "balance_loss_mlp": 1.02283251,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 2.326379567580216,
      "language_loss": 0.87924516,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.90159941,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 3.8132944107055664
    },
    {
      "auxiliary_loss_clip": 0.01206523,
      "auxiliary_loss_mlp": 0.01031882,
      "balance_loss_clip": 1.05994463,
      "balance_loss_mlp": 1.02247047,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 2.330961458624943,
      "language_loss": 0.753838,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.77622211,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 2.6998753547668457
    },
    {
      "auxiliary_loss_clip": 0.01205835,
      "auxiliary_loss_mlp": 0.01041649,
      "balance_loss_clip": 0.86999071,
      "balance_loss_mlp": 1.03167701,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 1.9525466876296995,
      "language_loss": 0.78972995,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.81220484,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 3.8152706623077393
    },
    {
      "auxiliary_loss_clip": 0.01203383,
      "auxiliary_loss_mlp": 0.01035484,
      "balance_loss_clip": 0.984532,
      "balance_loss_mlp": 1.02639413,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.6332275785119594,
      "language_loss": 0.74426943,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76665807,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 2.8196496963500977
    },
    {
      "auxiliary_loss_clip": 0.01201724,
      "auxiliary_loss_mlp": 0.01036914,
      "balance_loss_clip": 0.98510003,
      "balance_loss_mlp": 1.02729952,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 2.6326593186087197,
      "language_loss": 0.65299094,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.67537731,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 3.67911958694458
    },
    {
      "auxiliary_loss_clip": 0.01205319,
      "auxiliary_loss_mlp": 0.01032108,
      "balance_loss_clip": 0.98267645,
      "balance_loss_mlp": 1.02283382,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 1.9948356586058065,
      "language_loss": 0.81858945,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84096366,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.7684051990509033
    },
    {
      "auxiliary_loss_clip": 0.01212555,
      "auxiliary_loss_mlp": 0.01039679,
      "balance_loss_clip": 1.06330085,
      "balance_loss_mlp": 1.02993333,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.169740407732504,
      "language_loss": 0.91244185,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.93496418,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.7383079528808594
    },
    {
      "auxiliary_loss_clip": 0.01207858,
      "auxiliary_loss_mlp": 0.01035489,
      "balance_loss_clip": 1.02309072,
      "balance_loss_mlp": 1.02617896,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 1.7015354189568386,
      "language_loss": 0.79726261,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.81969607,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.8050103187561035
    },
    {
      "auxiliary_loss_clip": 0.01208302,
      "auxiliary_loss_mlp": 0.01031879,
      "balance_loss_clip": 1.02467871,
      "balance_loss_mlp": 1.02264071,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 1.810073029856845,
      "language_loss": 0.75258511,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77498692,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 2.8310413360595703
    },
    {
      "auxiliary_loss_clip": 0.01102926,
      "auxiliary_loss_mlp": 0.0112073,
      "balance_loss_clip": 0.95294297,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.8502954202547464,
      "language_loss": 0.69702321,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.7192598,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.362767457962036
    },
    {
      "auxiliary_loss_clip": 0.01204142,
      "auxiliary_loss_mlp": 0.0103477,
      "balance_loss_clip": 0.87062156,
      "balance_loss_mlp": 1.02534604,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.4927595166308207,
      "language_loss": 0.73787189,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.760261,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.8763625621795654
    },
    {
      "auxiliary_loss_clip": 0.01210141,
      "auxiliary_loss_mlp": 0.01034359,
      "balance_loss_clip": 1.02419424,
      "balance_loss_mlp": 1.02548981,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 2.0065799894966623,
      "language_loss": 0.65350997,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67595494,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 2.6943275928497314
    },
    {
      "auxiliary_loss_clip": 0.01206534,
      "auxiliary_loss_mlp": 0.01031886,
      "balance_loss_clip": 1.02346361,
      "balance_loss_mlp": 1.02248049,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 4.924604489561619,
      "language_loss": 0.74051905,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.76290321,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.7531888484954834
    },
    {
      "auxiliary_loss_clip": 0.01204352,
      "auxiliary_loss_mlp": 0.01041684,
      "balance_loss_clip": 0.98664862,
      "balance_loss_mlp": 1.03249335,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 3.055708789425653,
      "language_loss": 0.71404701,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.73650742,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.7513275146484375
    },
    {
      "auxiliary_loss_clip": 0.01207085,
      "auxiliary_loss_mlp": 0.01034795,
      "balance_loss_clip": 0.98604929,
      "balance_loss_mlp": 1.02482295,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 1.7567837552567662,
      "language_loss": 0.78601885,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.80843771,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.8303563594818115
    },
    {
      "auxiliary_loss_clip": 0.01205792,
      "auxiliary_loss_mlp": 0.01029315,
      "balance_loss_clip": 1.02343178,
      "balance_loss_mlp": 1.02006471,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 1.6794268230907894,
      "language_loss": 0.77808177,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.80043286,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 2.7931671142578125
    },
    {
      "auxiliary_loss_clip": 0.01198284,
      "auxiliary_loss_mlp": 0.01042029,
      "balance_loss_clip": 0.98291278,
      "balance_loss_mlp": 1.0322063,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 1.910203908738862,
      "language_loss": 0.87994659,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.90234971,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.7714200019836426
    },
    {
      "auxiliary_loss_clip": 0.01212252,
      "auxiliary_loss_mlp": 0.01034111,
      "balance_loss_clip": 1.06861126,
      "balance_loss_mlp": 1.02487278,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 1.5190230912051035,
      "language_loss": 0.78732371,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.80978739,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.864046335220337
    },
    {
      "auxiliary_loss_clip": 0.01209275,
      "auxiliary_loss_mlp": 0.0104062,
      "balance_loss_clip": 0.98275763,
      "balance_loss_mlp": 1.03078532,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 1.8248262896698275,
      "language_loss": 0.69483221,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71733117,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.9991114139556885
    },
    {
      "auxiliary_loss_clip": 0.01202966,
      "auxiliary_loss_mlp": 0.0112653,
      "balance_loss_clip": 0.98335713,
      "balance_loss_mlp": 0.0,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 1.8054731746260075,
      "language_loss": 0.82217205,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.84546697,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 2.8532121181488037
    },
    {
      "auxiliary_loss_clip": 0.01206912,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.06282282,
      "balance_loss_mlp": 1.0192976,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.4273998710823768,
      "language_loss": 0.84690142,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.86925375,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 2.7715067863464355
    },
    {
      "auxiliary_loss_clip": 0.011986,
      "auxiliary_loss_mlp": 0.01040075,
      "balance_loss_clip": 0.94312966,
      "balance_loss_mlp": 1.03091967,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 2.0898230806010814,
      "language_loss": 0.79375434,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.81614107,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.8703200817108154
    },
    {
      "auxiliary_loss_clip": 0.0120129,
      "auxiliary_loss_mlp": 0.01031158,
      "balance_loss_clip": 1.01824939,
      "balance_loss_mlp": 1.021788,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.324513850480093,
      "language_loss": 0.67901659,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.70134103,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.7726387977600098
    },
    {
      "auxiliary_loss_clip": 0.01196169,
      "auxiliary_loss_mlp": 0.01038399,
      "balance_loss_clip": 0.90306336,
      "balance_loss_mlp": 1.02860618,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.6159710428301837,
      "language_loss": 0.70680499,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.72915071,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 3.803751230239868
    },
    {
      "auxiliary_loss_clip": 0.01194223,
      "auxiliary_loss_mlp": 0.01038407,
      "balance_loss_clip": 0.86139834,
      "balance_loss_mlp": 1.02858472,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 1.9817820378281277,
      "language_loss": 0.74225807,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76458442,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 2.819650888442993
    },
    {
      "auxiliary_loss_clip": 0.01207149,
      "auxiliary_loss_mlp": 0.01034577,
      "balance_loss_clip": 0.98173583,
      "balance_loss_mlp": 1.02549887,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 1.9370489565816411,
      "language_loss": 0.81340176,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83581901,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.842355728149414
    },
    {
      "auxiliary_loss_clip": 0.01194695,
      "auxiliary_loss_mlp": 0.01029579,
      "balance_loss_clip": 0.98290682,
      "balance_loss_mlp": 1.01961339,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 1.9924123777032519,
      "language_loss": 0.72195804,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.74420083,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 3.775269031524658
    },
    {
      "auxiliary_loss_clip": 0.01204594,
      "auxiliary_loss_mlp": 0.01036836,
      "balance_loss_clip": 1.02406669,
      "balance_loss_mlp": 1.02760303,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 1.8258884439116658,
      "language_loss": 0.80558479,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82799911,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.7307796478271484
    },
    {
      "auxiliary_loss_clip": 0.01099777,
      "auxiliary_loss_mlp": 0.01007477,
      "balance_loss_clip": 1.02584171,
      "balance_loss_mlp": 1.00446129,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.782425265247041,
      "language_loss": 0.58146775,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60254025,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 4.144571542739868
    },
    {
      "auxiliary_loss_clip": 0.01208023,
      "auxiliary_loss_mlp": 0.01030808,
      "balance_loss_clip": 1.01985502,
      "balance_loss_mlp": 1.02209949,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 11.56305535777128,
      "language_loss": 0.94360435,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96599269,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 2.704148292541504
    },
    {
      "auxiliary_loss_clip": 0.0120694,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 0.98243183,
      "balance_loss_mlp": 1.02223945,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 2.6351740194886877,
      "language_loss": 0.78612912,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.80851942,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 3.688283920288086
    },
    {
      "auxiliary_loss_clip": 0.01203317,
      "auxiliary_loss_mlp": 0.01031392,
      "balance_loss_clip": 0.8720572,
      "balance_loss_mlp": 1.02181363,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 1.8096740431158096,
      "language_loss": 0.80534601,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.8276931,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.8836238384246826
    },
    {
      "auxiliary_loss_clip": 0.01208656,
      "auxiliary_loss_mlp": 0.01039513,
      "balance_loss_clip": 0.98564577,
      "balance_loss_mlp": 1.02997637,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.3931138988466034,
      "language_loss": 0.69397473,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.71645641,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.8993759155273438
    },
    {
      "auxiliary_loss_clip": 0.01205192,
      "auxiliary_loss_mlp": 0.01034177,
      "balance_loss_clip": 0.98528218,
      "balance_loss_mlp": 1.024593,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 1.9552447396442891,
      "language_loss": 0.89817834,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.92057198,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.7694194316864014
    },
    {
      "auxiliary_loss_clip": 0.0120382,
      "auxiliary_loss_mlp": 0.01126647,
      "balance_loss_clip": 0.90523785,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 1.7960155228369694,
      "language_loss": 0.83626759,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.85957229,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.8484573364257812
    },
    {
      "auxiliary_loss_clip": 0.01208244,
      "auxiliary_loss_mlp": 0.01039158,
      "balance_loss_clip": 0.98026162,
      "balance_loss_mlp": 1.02920389,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 2.080697222237044,
      "language_loss": 0.73689127,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.75936526,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 2.801345109939575
    },
    {
      "auxiliary_loss_clip": 0.0120008,
      "auxiliary_loss_mlp": 0.01032286,
      "balance_loss_clip": 0.94320369,
      "balance_loss_mlp": 1.02295232,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 1.9174088258450157,
      "language_loss": 0.8285501,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85087377,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.934187650680542
    },
    {
      "auxiliary_loss_clip": 0.0120499,
      "auxiliary_loss_mlp": 0.01034092,
      "balance_loss_clip": 1.01936269,
      "balance_loss_mlp": 1.02438807,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 1.9289038992986438,
      "language_loss": 0.75373513,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77612591,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.9695558547973633
    },
    {
      "auxiliary_loss_clip": 0.01202247,
      "auxiliary_loss_mlp": 0.01035748,
      "balance_loss_clip": 0.9393298,
      "balance_loss_mlp": 1.02643156,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 1.8316735893935578,
      "language_loss": 0.8668021,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.88918209,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.7849349975585938
    },
    {
      "auxiliary_loss_clip": 0.01214569,
      "auxiliary_loss_mlp": 0.0103626,
      "balance_loss_clip": 1.06163836,
      "balance_loss_mlp": 1.02571642,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 2.6699295356434107,
      "language_loss": 0.80535954,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82786781,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.797367572784424
    },
    {
      "auxiliary_loss_clip": 0.01099156,
      "auxiliary_loss_mlp": 0.01018132,
      "balance_loss_clip": 0.95097291,
      "balance_loss_mlp": 1.01488996,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.75049510380798,
      "language_loss": 0.6063484,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62752128,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.4091081619262695
    },
    {
      "auxiliary_loss_clip": 0.012083,
      "auxiliary_loss_mlp": 0.01026872,
      "balance_loss_clip": 0.94219851,
      "balance_loss_mlp": 1.01741898,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 1.7885440488632327,
      "language_loss": 0.72056484,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74291652,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.8384876251220703
    },
    {
      "auxiliary_loss_clip": 0.01184568,
      "auxiliary_loss_mlp": 0.01031206,
      "balance_loss_clip": 0.90247643,
      "balance_loss_mlp": 1.02168131,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 1.7839844618151754,
      "language_loss": 0.86523283,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.88739055,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 2.869396686553955
    },
    {
      "auxiliary_loss_clip": 0.01196113,
      "auxiliary_loss_mlp": 0.01036007,
      "balance_loss_clip": 0.98160881,
      "balance_loss_mlp": 1.02663684,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 1.8706762997156186,
      "language_loss": 0.80594575,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.82826698,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 2.7911384105682373
    },
    {
      "auxiliary_loss_clip": 0.01201632,
      "auxiliary_loss_mlp": 0.01034333,
      "balance_loss_clip": 1.01991224,
      "balance_loss_mlp": 1.02483165,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 2.0717942198747314,
      "language_loss": 0.80946755,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.83182716,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.745640516281128
    },
    {
      "auxiliary_loss_clip": 0.01206512,
      "auxiliary_loss_mlp": 0.01043403,
      "balance_loss_clip": 1.02225065,
      "balance_loss_mlp": 1.03402162,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.1464950854489793,
      "language_loss": 0.71780759,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.74030674,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 2.879973888397217
    },
    {
      "auxiliary_loss_clip": 0.01207368,
      "auxiliary_loss_mlp": 0.01039592,
      "balance_loss_clip": 0.94279045,
      "balance_loss_mlp": 1.03037667,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 2.2239903535208225,
      "language_loss": 0.66226274,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68473232,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 2.8318564891815186
    },
    {
      "auxiliary_loss_clip": 0.01208267,
      "auxiliary_loss_mlp": 0.01035122,
      "balance_loss_clip": 1.02510166,
      "balance_loss_mlp": 1.02560949,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 2.6154313409293075,
      "language_loss": 0.72930855,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.75174242,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.733745574951172
    },
    {
      "auxiliary_loss_clip": 0.01105386,
      "auxiliary_loss_mlp": 0.01012593,
      "balance_loss_clip": 0.98930323,
      "balance_loss_mlp": 1.00939798,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.2493146732827696,
      "language_loss": 0.68254048,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70372027,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.184849262237549
    },
    {
      "auxiliary_loss_clip": 0.01206397,
      "auxiliary_loss_mlp": 0.01036923,
      "balance_loss_clip": 0.94114774,
      "balance_loss_mlp": 1.02779758,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 1.8608339701512995,
      "language_loss": 0.76759899,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.79003215,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 3.809687376022339
    },
    {
      "auxiliary_loss_clip": 0.01187482,
      "auxiliary_loss_mlp": 0.01127173,
      "balance_loss_clip": 0.82591009,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 1.5888521668783364,
      "language_loss": 0.81066501,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83381152,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.828202247619629
    },
    {
      "auxiliary_loss_clip": 0.01201097,
      "auxiliary_loss_mlp": 0.01034983,
      "balance_loss_clip": 0.98216003,
      "balance_loss_mlp": 1.02479017,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.89758411094116,
      "language_loss": 0.79044056,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.81280136,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.6976194381713867
    },
    {
      "auxiliary_loss_clip": 0.01198076,
      "auxiliary_loss_mlp": 0.01032035,
      "balance_loss_clip": 0.97984254,
      "balance_loss_mlp": 1.02252793,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 2.9012136515718927,
      "language_loss": 0.7601403,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.78244138,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 3.674920082092285
    },
    {
      "auxiliary_loss_clip": 0.01204521,
      "auxiliary_loss_mlp": 0.01039366,
      "balance_loss_clip": 1.01930797,
      "balance_loss_mlp": 1.02957928,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 1.662839503412153,
      "language_loss": 0.80694288,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.8293817,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.7220659255981445
    },
    {
      "auxiliary_loss_clip": 0.01186131,
      "auxiliary_loss_mlp": 0.01126598,
      "balance_loss_clip": 0.93856561,
      "balance_loss_mlp": 0.0,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 5.357890082691454,
      "language_loss": 0.8146742,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.83780152,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 3.755364418029785
    },
    {
      "auxiliary_loss_clip": 0.01210773,
      "auxiliary_loss_mlp": 0.01033645,
      "balance_loss_clip": 1.06107318,
      "balance_loss_mlp": 1.02372658,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.0160104188669847,
      "language_loss": 0.90203732,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.92448151,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.8083724975585938
    },
    {
      "auxiliary_loss_clip": 0.01193805,
      "auxiliary_loss_mlp": 0.01036299,
      "balance_loss_clip": 0.89934599,
      "balance_loss_mlp": 1.02617264,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.8610533670624425,
      "language_loss": 0.65804124,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.68034232,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 3.769620180130005
    },
    {
      "auxiliary_loss_clip": 0.01189516,
      "auxiliary_loss_mlp": 0.01043787,
      "balance_loss_clip": 0.9035629,
      "balance_loss_mlp": 1.03437543,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 1.8714172958054678,
      "language_loss": 0.89559484,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.9179278,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.8648366928100586
    },
    {
      "auxiliary_loss_clip": 0.01206225,
      "auxiliary_loss_mlp": 0.01034694,
      "balance_loss_clip": 0.9806428,
      "balance_loss_mlp": 1.02461505,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.2790550011212236,
      "language_loss": 0.79243112,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.81484026,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.8190042972564697
    },
    {
      "auxiliary_loss_clip": 0.0120004,
      "auxiliary_loss_mlp": 0.01033937,
      "balance_loss_clip": 0.94409776,
      "balance_loss_mlp": 1.02473426,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.0431742727291273,
      "language_loss": 0.9232288,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.94556856,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.7222740650177
    },
    {
      "auxiliary_loss_clip": 0.01194264,
      "auxiliary_loss_mlp": 0.01037887,
      "balance_loss_clip": 1.0188843,
      "balance_loss_mlp": 1.02782619,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 1.897063826943328,
      "language_loss": 0.80782264,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.83014417,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.671287775039673
    },
    {
      "auxiliary_loss_clip": 0.01103236,
      "auxiliary_loss_mlp": 0.01009765,
      "balance_loss_clip": 0.98841751,
      "balance_loss_mlp": 1.00652206,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.7941316031030269,
      "language_loss": 0.57033277,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.59146273,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 3.2953004837036133
    },
    {
      "auxiliary_loss_clip": 0.01200534,
      "auxiliary_loss_mlp": 0.01032674,
      "balance_loss_clip": 0.90553093,
      "balance_loss_mlp": 1.02338767,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 1.7311325650041343,
      "language_loss": 0.85304368,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87537575,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 2.734788656234741
    },
    {
      "auxiliary_loss_clip": 0.01188192,
      "auxiliary_loss_mlp": 0.01037157,
      "balance_loss_clip": 0.9782905,
      "balance_loss_mlp": 1.02805519,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 2.044157064328177,
      "language_loss": 0.82792872,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85018224,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.744917392730713
    },
    {
      "auxiliary_loss_clip": 0.01189445,
      "auxiliary_loss_mlp": 0.01036964,
      "balance_loss_clip": 0.94112146,
      "balance_loss_mlp": 1.02751064,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 1.7428750773227486,
      "language_loss": 0.80249923,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.8247633,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.746647357940674
    },
    {
      "auxiliary_loss_clip": 0.01207679,
      "auxiliary_loss_mlp": 0.0103543,
      "balance_loss_clip": 1.02027631,
      "balance_loss_mlp": 1.02564931,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 1.9540066677087273,
      "language_loss": 0.84892339,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.87135446,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.8453307151794434
    },
    {
      "auxiliary_loss_clip": 0.01201221,
      "auxiliary_loss_mlp": 0.0103345,
      "balance_loss_clip": 1.02010059,
      "balance_loss_mlp": 1.02359164,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.0068311874221525,
      "language_loss": 0.61964762,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.64199436,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.7226016521453857
    },
    {
      "auxiliary_loss_clip": 0.01205975,
      "auxiliary_loss_mlp": 0.01033313,
      "balance_loss_clip": 0.98386425,
      "balance_loss_mlp": 1.02369881,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 2.1000884118270786,
      "language_loss": 0.78315818,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80555105,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.7706358432769775
    },
    {
      "auxiliary_loss_clip": 0.01200976,
      "auxiliary_loss_mlp": 0.010324,
      "balance_loss_clip": 0.94274759,
      "balance_loss_mlp": 1.02270222,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 2.1841494688133203,
      "language_loss": 0.86197567,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.88430947,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.7751827239990234
    },
    {
      "auxiliary_loss_clip": 0.01190619,
      "auxiliary_loss_mlp": 0.01035521,
      "balance_loss_clip": 0.94131845,
      "balance_loss_mlp": 1.02582943,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 1.913803660922693,
      "language_loss": 0.70860887,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.73087025,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.847914457321167
    },
    {
      "auxiliary_loss_clip": 0.01206966,
      "auxiliary_loss_mlp": 0.01037272,
      "balance_loss_clip": 1.02420068,
      "balance_loss_mlp": 1.0275743,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 1.5972738179154966,
      "language_loss": 0.70321119,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72565353,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 2.734219789505005
    },
    {
      "auxiliary_loss_clip": 0.01197375,
      "auxiliary_loss_mlp": 0.01032496,
      "balance_loss_clip": 0.9786917,
      "balance_loss_mlp": 1.02283406,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 1.8493573838633472,
      "language_loss": 0.78949034,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.81178904,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 2.8467023372650146
    },
    {
      "auxiliary_loss_clip": 0.01204291,
      "auxiliary_loss_mlp": 0.01035129,
      "balance_loss_clip": 0.90589929,
      "balance_loss_mlp": 1.02511537,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 2.962633374181734,
      "language_loss": 0.8421725,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.86456674,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.825985908508301
    },
    {
      "auxiliary_loss_clip": 0.01206672,
      "auxiliary_loss_mlp": 0.01032775,
      "balance_loss_clip": 1.06054473,
      "balance_loss_mlp": 1.0229938,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 1.790533668062006,
      "language_loss": 0.89302027,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.91541469,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.710679769515991
    },
    {
      "auxiliary_loss_clip": 0.01203425,
      "auxiliary_loss_mlp": 0.01032079,
      "balance_loss_clip": 1.02032351,
      "balance_loss_mlp": 1.02298331,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 1.5422917978635073,
      "language_loss": 0.85691297,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.87926805,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.8056585788726807
    },
    {
      "auxiliary_loss_clip": 0.01181667,
      "auxiliary_loss_mlp": 0.01039197,
      "balance_loss_clip": 0.97708297,
      "balance_loss_mlp": 1.02971399,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 1.8668187987972122,
      "language_loss": 0.76907825,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.79128683,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 3.7492730617523193
    },
    {
      "auxiliary_loss_clip": 0.01205123,
      "auxiliary_loss_mlp": 0.01036525,
      "balance_loss_clip": 1.05943656,
      "balance_loss_mlp": 1.02684569,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 2.776014157410966,
      "language_loss": 0.76720721,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.78962368,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 2.64748215675354
    },
    {
      "auxiliary_loss_clip": 0.01203037,
      "auxiliary_loss_mlp": 0.01033448,
      "balance_loss_clip": 0.90748715,
      "balance_loss_mlp": 1.02467966,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 1.8555533173559013,
      "language_loss": 0.78030109,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.80266595,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.7963924407958984
    },
    {
      "auxiliary_loss_clip": 0.01206937,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.05971205,
      "balance_loss_mlp": 1.02098525,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 4.315560577383983,
      "language_loss": 0.77414197,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.79651994,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 3.698664903640747
    },
    {
      "auxiliary_loss_clip": 0.01208611,
      "auxiliary_loss_mlp": 0.01034529,
      "balance_loss_clip": 0.94431925,
      "balance_loss_mlp": 1.02450943,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 2.021452535811851,
      "language_loss": 0.79639584,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81882727,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.8130719661712646
    },
    {
      "auxiliary_loss_clip": 0.01195263,
      "auxiliary_loss_mlp": 0.01031535,
      "balance_loss_clip": 0.93968493,
      "balance_loss_mlp": 1.0221169,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 3.140744780477653,
      "language_loss": 0.73532116,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.75758922,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 2.746267080307007
    },
    {
      "auxiliary_loss_clip": 0.01198093,
      "auxiliary_loss_mlp": 0.0103566,
      "balance_loss_clip": 0.9416917,
      "balance_loss_mlp": 1.02592063,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 4.233732726916911,
      "language_loss": 0.72752941,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.74986696,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 3.7490127086639404
    },
    {
      "auxiliary_loss_clip": 0.01202502,
      "auxiliary_loss_mlp": 0.01036757,
      "balance_loss_clip": 0.98228592,
      "balance_loss_mlp": 1.02742314,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 2.8339990539447792,
      "language_loss": 0.83606398,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.85845655,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 3.692861318588257
    },
    {
      "auxiliary_loss_clip": 0.01207989,
      "auxiliary_loss_mlp": 0.0103298,
      "balance_loss_clip": 0.90421069,
      "balance_loss_mlp": 1.02322841,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 2.5146872357936405,
      "language_loss": 0.54177457,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.56418425,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.740969657897949
    },
    {
      "auxiliary_loss_clip": 0.01203503,
      "auxiliary_loss_mlp": 0.01039599,
      "balance_loss_clip": 1.02347076,
      "balance_loss_mlp": 1.03006792,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.4096695815022127,
      "language_loss": 0.75048614,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77291715,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 2.7696175575256348
    },
    {
      "auxiliary_loss_clip": 0.01210782,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.02172899,
      "balance_loss_mlp": 1.02401614,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 2.1641636656604,
      "language_loss": 0.63629663,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.6587435,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.635300874710083
    },
    {
      "auxiliary_loss_clip": 0.01207212,
      "auxiliary_loss_mlp": 0.01033014,
      "balance_loss_clip": 0.98262805,
      "balance_loss_mlp": 1.02297091,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 1.9208888004294757,
      "language_loss": 0.85416275,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87656498,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.7218565940856934
    },
    {
      "auxiliary_loss_clip": 0.01202447,
      "auxiliary_loss_mlp": 0.0103277,
      "balance_loss_clip": 0.98132026,
      "balance_loss_mlp": 1.02401423,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 1.6986524338791158,
      "language_loss": 0.87419897,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.89655113,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 2.740952968597412
    },
    {
      "auxiliary_loss_clip": 0.01200725,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 0.8682493,
      "balance_loss_mlp": 1.02090728,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 2.0457535931584134,
      "language_loss": 0.92741776,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94972634,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 2.7873501777648926
    },
    {
      "auxiliary_loss_clip": 0.01204784,
      "auxiliary_loss_mlp": 0.01033249,
      "balance_loss_clip": 0.94169861,
      "balance_loss_mlp": 1.0238142,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 2.0858937586322046,
      "language_loss": 0.88036919,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.90274942,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.7589240074157715
    },
    {
      "auxiliary_loss_clip": 0.01213412,
      "auxiliary_loss_mlp": 0.01036986,
      "balance_loss_clip": 0.98621356,
      "balance_loss_mlp": 1.02718127,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 1.8587776220229726,
      "language_loss": 0.83590543,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.8584094,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.695981502532959
    },
    {
      "auxiliary_loss_clip": 0.01212878,
      "auxiliary_loss_mlp": 0.01038863,
      "balance_loss_clip": 1.01956499,
      "balance_loss_mlp": 1.02838433,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 1.633875239145064,
      "language_loss": 0.82724297,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.84976035,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.714031934738159
    },
    {
      "auxiliary_loss_clip": 0.01203648,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 0.98214352,
      "balance_loss_mlp": 1.02133322,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 4.435341867021555,
      "language_loss": 0.78424478,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80659389,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.7070724964141846
    },
    {
      "auxiliary_loss_clip": 0.01209938,
      "auxiliary_loss_mlp": 0.01038229,
      "balance_loss_clip": 1.02059269,
      "balance_loss_mlp": 1.02803111,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 1.9430520891192118,
      "language_loss": 0.8280046,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.85048628,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.663074493408203
    },
    {
      "auxiliary_loss_clip": 0.01192534,
      "auxiliary_loss_mlp": 0.01032334,
      "balance_loss_clip": 0.97738695,
      "balance_loss_mlp": 1.02273726,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 1.9588566291031908,
      "language_loss": 0.84558666,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.8678354,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.6334869861602783
    },
    {
      "auxiliary_loss_clip": 0.01203661,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.02146101,
      "balance_loss_mlp": 1.02263141,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.5325928252574272,
      "language_loss": 0.8380546,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.86040372,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.719998836517334
    },
    {
      "auxiliary_loss_clip": 0.01203827,
      "auxiliary_loss_mlp": 0.01035362,
      "balance_loss_clip": 0.94143552,
      "balance_loss_mlp": 1.02522993,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 2.3454314032624555,
      "language_loss": 0.74910057,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77149248,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 2.927945375442505
    },
    {
      "auxiliary_loss_clip": 0.01197494,
      "auxiliary_loss_mlp": 0.01033817,
      "balance_loss_clip": 0.94631141,
      "balance_loss_mlp": 1.02421451,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 2.3239525532302525,
      "language_loss": 0.8307991,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85311222,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.753730297088623
    },
    {
      "auxiliary_loss_clip": 0.01190024,
      "auxiliary_loss_mlp": 0.01032742,
      "balance_loss_clip": 1.01468849,
      "balance_loss_mlp": 1.02313328,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 1.9895444484857585,
      "language_loss": 0.68316048,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70538819,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 2.722412586212158
    },
    {
      "auxiliary_loss_clip": 0.01210616,
      "auxiliary_loss_mlp": 0.01036182,
      "balance_loss_clip": 1.02332854,
      "balance_loss_mlp": 1.02632928,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 1.817911789469092,
      "language_loss": 0.71095902,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.73342699,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.9237892627716064
    },
    {
      "auxiliary_loss_clip": 0.01201672,
      "auxiliary_loss_mlp": 0.01036986,
      "balance_loss_clip": 0.98028237,
      "balance_loss_mlp": 1.02731788,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 1.8412963333429704,
      "language_loss": 0.85318178,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.87556839,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.775118350982666
    },
    {
      "auxiliary_loss_clip": 0.01196606,
      "auxiliary_loss_mlp": 0.01126449,
      "balance_loss_clip": 1.01929891,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.2637437968360032,
      "language_loss": 0.67618024,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.6994108,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 4.205892562866211
    },
    {
      "auxiliary_loss_clip": 0.01199087,
      "auxiliary_loss_mlp": 0.01031035,
      "balance_loss_clip": 0.97976923,
      "balance_loss_mlp": 1.02237439,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.683918624744205,
      "language_loss": 0.82186049,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84416169,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.8536882400512695
    },
    {
      "auxiliary_loss_clip": 0.01206582,
      "auxiliary_loss_mlp": 0.01035461,
      "balance_loss_clip": 1.05956864,
      "balance_loss_mlp": 1.02579284,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 2.214202045897326,
      "language_loss": 0.75643522,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.77885562,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.8227498531341553
    },
    {
      "auxiliary_loss_clip": 0.01193942,
      "auxiliary_loss_mlp": 0.01034277,
      "balance_loss_clip": 0.94000459,
      "balance_loss_mlp": 1.02451408,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 2.1878680204599035,
      "language_loss": 0.71915132,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.7414335,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.7837016582489014
    },
    {
      "auxiliary_loss_clip": 0.01196786,
      "auxiliary_loss_mlp": 0.01038402,
      "balance_loss_clip": 0.93770528,
      "balance_loss_mlp": 1.02943122,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 2.260668189011596,
      "language_loss": 0.9113239,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.93367577,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 3.6098275184631348
    },
    {
      "auxiliary_loss_clip": 0.01187887,
      "auxiliary_loss_mlp": 0.01033279,
      "balance_loss_clip": 0.86256588,
      "balance_loss_mlp": 1.02306843,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 2.1470381215072645,
      "language_loss": 0.76351786,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.78572947,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.773604154586792
    },
    {
      "auxiliary_loss_clip": 0.0119437,
      "auxiliary_loss_mlp": 0.01026387,
      "balance_loss_clip": 0.94313085,
      "balance_loss_mlp": 1.01761878,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 1.678465778283045,
      "language_loss": 0.83332217,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.85552979,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 3.7224810123443604
    },
    {
      "auxiliary_loss_clip": 0.01208788,
      "auxiliary_loss_mlp": 0.01125979,
      "balance_loss_clip": 1.06193781,
      "balance_loss_mlp": 0.0,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 1.9737019634561825,
      "language_loss": 0.82118243,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.84453011,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 3.5307416915893555
    },
    {
      "auxiliary_loss_clip": 0.01205945,
      "auxiliary_loss_mlp": 0.01037804,
      "balance_loss_clip": 0.98442036,
      "balance_loss_mlp": 1.0282793,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 1.9397081092763333,
      "language_loss": 0.75992703,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.78236449,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 2.8267276287078857
    },
    {
      "auxiliary_loss_clip": 0.01206201,
      "auxiliary_loss_mlp": 0.01031291,
      "balance_loss_clip": 1.02078521,
      "balance_loss_mlp": 1.02218962,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 1.9546801490206054,
      "language_loss": 0.86082965,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.88320464,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.652449131011963
    },
    {
      "auxiliary_loss_clip": 0.01202436,
      "auxiliary_loss_mlp": 0.01034336,
      "balance_loss_clip": 1.02139735,
      "balance_loss_mlp": 1.02567506,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 1.6056510185274333,
      "language_loss": 0.80815578,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83052349,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.698108673095703
    },
    {
      "auxiliary_loss_clip": 0.01211229,
      "auxiliary_loss_mlp": 0.01032834,
      "balance_loss_clip": 1.02171481,
      "balance_loss_mlp": 1.02345777,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 2.139862749318847,
      "language_loss": 0.68464369,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.7070843,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.7305123805999756
    },
    {
      "auxiliary_loss_clip": 0.01203396,
      "auxiliary_loss_mlp": 0.01034513,
      "balance_loss_clip": 0.86550504,
      "balance_loss_mlp": 1.02571499,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 1.7997450749248542,
      "language_loss": 0.82411921,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.84649825,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.8618040084838867
    },
    {
      "auxiliary_loss_clip": 0.01118949,
      "auxiliary_loss_mlp": 0.01003018,
      "balance_loss_clip": 0.92382199,
      "balance_loss_mlp": 1.00027597,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 1.1266226278611848,
      "language_loss": 0.64727449,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66849416,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.201826572418213
    },
    {
      "auxiliary_loss_clip": 0.01204087,
      "auxiliary_loss_mlp": 0.01032115,
      "balance_loss_clip": 1.01912141,
      "balance_loss_mlp": 1.02300167,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 1.8719312651348474,
      "language_loss": 0.86606431,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.88842642,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 2.6548044681549072
    },
    {
      "auxiliary_loss_clip": 0.01203482,
      "auxiliary_loss_mlp": 0.01039559,
      "balance_loss_clip": 1.0598886,
      "balance_loss_mlp": 1.03095794,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 1.7607677470068772,
      "language_loss": 0.75337654,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77580696,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.702402114868164
    },
    {
      "auxiliary_loss_clip": 0.01199768,
      "auxiliary_loss_mlp": 0.01035998,
      "balance_loss_clip": 1.01810765,
      "balance_loss_mlp": 1.02706909,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 2.3929544054630982,
      "language_loss": 0.73545736,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.757815,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.7088730335235596
    },
    {
      "auxiliary_loss_clip": 0.01197449,
      "auxiliary_loss_mlp": 0.0103492,
      "balance_loss_clip": 0.9777106,
      "balance_loss_mlp": 1.02581835,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 2.2526585376363233,
      "language_loss": 0.81513155,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.83745527,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.7330660820007324
    },
    {
      "auxiliary_loss_clip": 0.012052,
      "auxiliary_loss_mlp": 0.01125942,
      "balance_loss_clip": 1.05954361,
      "balance_loss_mlp": 0.0,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 1.8687607100077,
      "language_loss": 0.83606088,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.85937238,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.640075206756592
    },
    {
      "auxiliary_loss_clip": 0.01193536,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 0.97891653,
      "balance_loss_mlp": 1.01935613,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 1.5424381660185806,
      "language_loss": 0.89537895,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91759551,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.7241549491882324
    },
    {
      "auxiliary_loss_clip": 0.01199623,
      "auxiliary_loss_mlp": 0.01036241,
      "balance_loss_clip": 0.98188657,
      "balance_loss_mlp": 1.02661514,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 2.045466212008437,
      "language_loss": 0.71738315,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.73974174,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.8514676094055176
    },
    {
      "auxiliary_loss_clip": 0.01185017,
      "auxiliary_loss_mlp": 0.01035295,
      "balance_loss_clip": 0.90681696,
      "balance_loss_mlp": 1.02600253,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.7328963503466122,
      "language_loss": 0.80859566,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.83079886,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 2.7943921089172363
    },
    {
      "auxiliary_loss_clip": 0.01211101,
      "auxiliary_loss_mlp": 0.01040094,
      "balance_loss_clip": 1.02414203,
      "balance_loss_mlp": 1.03134966,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.4926614864873184,
      "language_loss": 0.81936944,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.84188139,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.6943399906158447
    },
    {
      "auxiliary_loss_clip": 0.01209826,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 0.98329657,
      "balance_loss_mlp": 1.02191615,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 2.4446705483028963,
      "language_loss": 0.92906272,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.95146859,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.6943278312683105
    },
    {
      "auxiliary_loss_clip": 0.01206214,
      "auxiliary_loss_mlp": 0.01031029,
      "balance_loss_clip": 1.06024933,
      "balance_loss_mlp": 1.02168918,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 2.037712314232414,
      "language_loss": 0.70656323,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72893572,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.6207761764526367
    },
    {
      "auxiliary_loss_clip": 0.01202637,
      "auxiliary_loss_mlp": 0.01046696,
      "balance_loss_clip": 1.02439713,
      "balance_loss_mlp": 1.0374279,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 2.5327613849925354,
      "language_loss": 0.75854057,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.78103393,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 2.7032575607299805
    },
    {
      "auxiliary_loss_clip": 0.01205639,
      "auxiliary_loss_mlp": 0.01034092,
      "balance_loss_clip": 1.05960906,
      "balance_loss_mlp": 1.02405477,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.29608399965473,
      "language_loss": 0.83388495,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.85628223,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.703402280807495
    },
    {
      "auxiliary_loss_clip": 0.01203544,
      "auxiliary_loss_mlp": 0.01037691,
      "balance_loss_clip": 1.01992011,
      "balance_loss_mlp": 1.02857161,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 2.2246780300790667,
      "language_loss": 0.74610925,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76852161,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 3.781247138977051
    },
    {
      "auxiliary_loss_clip": 0.01185965,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 0.86138427,
      "balance_loss_mlp": 1.02020526,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.0710232676035316,
      "language_loss": 0.62310183,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.6452539,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 2.92177677154541
    },
    {
      "auxiliary_loss_clip": 0.01190033,
      "auxiliary_loss_mlp": 0.01030873,
      "balance_loss_clip": 0.90404987,
      "balance_loss_mlp": 1.0221765,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 3.6442356763263186,
      "language_loss": 0.84303808,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86524713,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 3.724660634994507
    },
    {
      "auxiliary_loss_clip": 0.01196185,
      "auxiliary_loss_mlp": 0.01030198,
      "balance_loss_clip": 0.94375461,
      "balance_loss_mlp": 1.02107263,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 2.0690843509884616,
      "language_loss": 0.83110082,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.85336465,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 2.7493810653686523
    },
    {
      "auxiliary_loss_clip": 0.01197957,
      "auxiliary_loss_mlp": 0.0103287,
      "balance_loss_clip": 0.94487607,
      "balance_loss_mlp": 1.02366066,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 2.1366878509376943,
      "language_loss": 0.83878565,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.86109394,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.705498456954956
    },
    {
      "auxiliary_loss_clip": 0.01201733,
      "auxiliary_loss_mlp": 0.01033255,
      "balance_loss_clip": 1.02199566,
      "balance_loss_mlp": 1.02395725,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 2.4154334890505518,
      "language_loss": 0.75632328,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.77867317,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 3.685453414916992
    },
    {
      "auxiliary_loss_clip": 0.01187649,
      "auxiliary_loss_mlp": 0.0103483,
      "balance_loss_clip": 0.97922397,
      "balance_loss_mlp": 1.02548361,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 1.8731725794791765,
      "language_loss": 0.84137189,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.86359674,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 3.623870372772217
    },
    {
      "auxiliary_loss_clip": 0.01195346,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.01919651,
      "balance_loss_mlp": 1.01827312,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 2.1343064189498957,
      "language_loss": 0.76710796,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78933519,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.6296117305755615
    },
    {
      "auxiliary_loss_clip": 0.01198944,
      "auxiliary_loss_mlp": 0.0103632,
      "balance_loss_clip": 0.98315489,
      "balance_loss_mlp": 1.0278089,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 2.2253053238613196,
      "language_loss": 0.78723848,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.80959117,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 2.6418826580047607
    },
    {
      "auxiliary_loss_clip": 0.0120035,
      "auxiliary_loss_mlp": 0.01034319,
      "balance_loss_clip": 0.94058061,
      "balance_loss_mlp": 1.02598059,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.4064792427763586,
      "language_loss": 0.70824325,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73058999,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.7750136852264404
    },
    {
      "auxiliary_loss_clip": 0.01198346,
      "auxiliary_loss_mlp": 0.01030982,
      "balance_loss_clip": 0.94381565,
      "balance_loss_mlp": 1.02248788,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 1.8549627558703807,
      "language_loss": 0.8662951,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.88858831,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.7396252155303955
    },
    {
      "auxiliary_loss_clip": 0.01110148,
      "auxiliary_loss_mlp": 0.01001758,
      "balance_loss_clip": 0.99787945,
      "balance_loss_mlp": 0.99888462,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9492952517108594,
      "language_loss": 0.62437803,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64549708,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.3745508193969727
    },
    {
      "auxiliary_loss_clip": 0.01204432,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.01942205,
      "balance_loss_mlp": 1.02257133,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 4.062011407758059,
      "language_loss": 0.71276677,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.73513055,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 2.635171890258789
    },
    {
      "auxiliary_loss_clip": 0.01199159,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 1.02042556,
      "balance_loss_mlp": 1.02118957,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 4.176897819375736,
      "language_loss": 0.84344918,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.86573929,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 2.706068754196167
    },
    {
      "auxiliary_loss_clip": 0.01191728,
      "auxiliary_loss_mlp": 0.0104214,
      "balance_loss_clip": 0.97928071,
      "balance_loss_mlp": 1.0330677,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 2.5004566800792065,
      "language_loss": 0.91110086,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.93343961,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.7195515632629395
    },
    {
      "auxiliary_loss_clip": 0.01204775,
      "auxiliary_loss_mlp": 0.0103478,
      "balance_loss_clip": 0.98570335,
      "balance_loss_mlp": 1.02514768,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.40301043301708,
      "language_loss": 0.85099769,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.87339324,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.705413341522217
    },
    {
      "auxiliary_loss_clip": 0.01205447,
      "auxiliary_loss_mlp": 0.0112536,
      "balance_loss_clip": 1.06098986,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 2.0995941933331297,
      "language_loss": 0.90633404,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.92964214,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.658223867416382
    },
    {
      "auxiliary_loss_clip": 0.01198034,
      "auxiliary_loss_mlp": 0.01033621,
      "balance_loss_clip": 1.02104282,
      "balance_loss_mlp": 1.02429914,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 3.0604406662484487,
      "language_loss": 0.77896667,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.80128324,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.6402950286865234
    },
    {
      "auxiliary_loss_clip": 0.01205578,
      "auxiliary_loss_mlp": 0.01029811,
      "balance_loss_clip": 1.06014967,
      "balance_loss_mlp": 1.02082264,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 2.224032966905301,
      "language_loss": 0.91512859,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.93748248,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.7073590755462646
    },
    {
      "auxiliary_loss_clip": 0.01203605,
      "auxiliary_loss_mlp": 0.0104384,
      "balance_loss_clip": 0.98403895,
      "balance_loss_mlp": 1.03441691,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 5.73240360066121,
      "language_loss": 0.7363233,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.75879776,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.684750556945801
    },
    {
      "auxiliary_loss_clip": 0.01202936,
      "auxiliary_loss_mlp": 0.01040195,
      "balance_loss_clip": 1.01994801,
      "balance_loss_mlp": 1.03063428,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 1.9277411131622397,
      "language_loss": 0.78943706,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.81186837,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 2.7303850650787354
    },
    {
      "auxiliary_loss_clip": 0.01183465,
      "auxiliary_loss_mlp": 0.01032424,
      "balance_loss_clip": 0.9399789,
      "balance_loss_mlp": 1.0236268,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 4.015439821689167,
      "language_loss": 0.78377676,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.80593568,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.6993701457977295
    },
    {
      "auxiliary_loss_clip": 0.01200672,
      "auxiliary_loss_mlp": 0.01034405,
      "balance_loss_clip": 1.02190566,
      "balance_loss_mlp": 1.0254643,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 1.9820609799723872,
      "language_loss": 0.8547979,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.87714863,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 2.802813768386841
    },
    {
      "auxiliary_loss_clip": 0.01205813,
      "auxiliary_loss_mlp": 0.01034625,
      "balance_loss_clip": 0.98184776,
      "balance_loss_mlp": 1.02582133,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 1.9766911700340488,
      "language_loss": 0.82663321,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.84903765,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.726977825164795
    },
    {
      "auxiliary_loss_clip": 0.01204698,
      "auxiliary_loss_mlp": 0.01125371,
      "balance_loss_clip": 1.06165874,
      "balance_loss_mlp": 0.0,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.596154639669897,
      "language_loss": 0.73801506,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.76131576,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.7225866317749023
    },
    {
      "auxiliary_loss_clip": 0.01190512,
      "auxiliary_loss_mlp": 0.01032006,
      "balance_loss_clip": 0.94038177,
      "balance_loss_mlp": 1.02280343,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 2.019715917040385,
      "language_loss": 0.67993283,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70215797,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 3.7701423168182373
    },
    {
      "auxiliary_loss_clip": 0.01210706,
      "auxiliary_loss_mlp": 0.01036446,
      "balance_loss_clip": 1.05957544,
      "balance_loss_mlp": 1.02648592,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.204439855528983,
      "language_loss": 0.68848032,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71095181,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 2.679001569747925
    },
    {
      "auxiliary_loss_clip": 0.01098986,
      "auxiliary_loss_mlp": 0.01006491,
      "balance_loss_clip": 0.91046256,
      "balance_loss_mlp": 1.00370169,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.8836634804989233,
      "language_loss": 0.5559774,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57703215,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.270258665084839
    },
    {
      "auxiliary_loss_clip": 0.01196605,
      "auxiliary_loss_mlp": 0.01031819,
      "balance_loss_clip": 0.97838187,
      "balance_loss_mlp": 1.02243078,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 1.853307281025987,
      "language_loss": 0.79069757,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.81298184,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 2.7274436950683594
    },
    {
      "auxiliary_loss_clip": 0.0119427,
      "auxiliary_loss_mlp": 0.01036878,
      "balance_loss_clip": 0.9456166,
      "balance_loss_mlp": 1.02818751,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 1.49105590308866,
      "language_loss": 0.82073045,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.8430419,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 3.713630199432373
    },
    {
      "auxiliary_loss_clip": 0.01096935,
      "auxiliary_loss_mlp": 0.01007685,
      "balance_loss_clip": 0.90980041,
      "balance_loss_mlp": 1.00474095,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.833116175910257,
      "language_loss": 0.6000728,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.62111902,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.369288206100464
    },
    {
      "auxiliary_loss_clip": 0.01189116,
      "auxiliary_loss_mlp": 0.01032714,
      "balance_loss_clip": 0.97975039,
      "balance_loss_mlp": 1.0236783,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 1.729226987635877,
      "language_loss": 0.70376563,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72598398,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 3.772744655609131
    },
    {
      "auxiliary_loss_clip": 0.01204101,
      "auxiliary_loss_mlp": 0.01032359,
      "balance_loss_clip": 1.02489042,
      "balance_loss_mlp": 1.02313209,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 1.9542427853048498,
      "language_loss": 0.83341467,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.85577929,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 3.7222962379455566
    },
    {
      "auxiliary_loss_clip": 0.01205651,
      "auxiliary_loss_mlp": 0.01034214,
      "balance_loss_clip": 0.94430631,
      "balance_loss_mlp": 1.02536845,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 2.3941731077986335,
      "language_loss": 0.80126631,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82366502,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.8246119022369385
    },
    {
      "auxiliary_loss_clip": 0.01197638,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 0.94564551,
      "balance_loss_mlp": 1.01898873,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.151917577656584,
      "language_loss": 0.78387165,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.80612135,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 2.80025577545166
    },
    {
      "auxiliary_loss_clip": 0.01199905,
      "auxiliary_loss_mlp": 0.01032708,
      "balance_loss_clip": 1.01831985,
      "balance_loss_mlp": 1.02303433,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 2.7254266433149086,
      "language_loss": 0.66880357,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69112968,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.658466100692749
    },
    {
      "auxiliary_loss_clip": 0.01198323,
      "auxiliary_loss_mlp": 0.01031147,
      "balance_loss_clip": 0.98184073,
      "balance_loss_mlp": 1.02186024,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 1.7817093381438038,
      "language_loss": 0.69625437,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.71854907,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.6657474040985107
    },
    {
      "auxiliary_loss_clip": 0.01202234,
      "auxiliary_loss_mlp": 0.01037321,
      "balance_loss_clip": 1.02198291,
      "balance_loss_mlp": 1.0277071,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 1.8171454361945936,
      "language_loss": 0.77495182,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79734737,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.6761014461517334
    },
    {
      "auxiliary_loss_clip": 0.01188899,
      "auxiliary_loss_mlp": 0.01034029,
      "balance_loss_clip": 0.97830713,
      "balance_loss_mlp": 1.02526093,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 1.7835600168804389,
      "language_loss": 0.65478384,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.67701316,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.642688035964966
    },
    {
      "auxiliary_loss_clip": 0.01187227,
      "auxiliary_loss_mlp": 0.01037086,
      "balance_loss_clip": 0.90148419,
      "balance_loss_mlp": 1.02785349,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 1.7600100524831097,
      "language_loss": 0.7800808,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.80232388,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 2.800309658050537
    },
    {
      "auxiliary_loss_clip": 0.01197516,
      "auxiliary_loss_mlp": 0.01035986,
      "balance_loss_clip": 1.01999092,
      "balance_loss_mlp": 1.02706909,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.124257777336987,
      "language_loss": 0.72899973,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75133473,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 2.728722095489502
    },
    {
      "auxiliary_loss_clip": 0.01203103,
      "auxiliary_loss_mlp": 0.01034402,
      "balance_loss_clip": 1.05802476,
      "balance_loss_mlp": 1.02568221,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.285822695229735,
      "language_loss": 0.74893665,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.77131164,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.556746244430542
    },
    {
      "auxiliary_loss_clip": 0.01205778,
      "auxiliary_loss_mlp": 0.01034644,
      "balance_loss_clip": 0.90838438,
      "balance_loss_mlp": 1.02550638,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 2.0333809223822708,
      "language_loss": 0.77478653,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.79719079,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 2.8218865394592285
    },
    {
      "auxiliary_loss_clip": 0.01198902,
      "auxiliary_loss_mlp": 0.01043426,
      "balance_loss_clip": 0.97864604,
      "balance_loss_mlp": 1.03419924,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 1.677297984988179,
      "language_loss": 0.80714118,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.82956451,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.738102912902832
    },
    {
      "auxiliary_loss_clip": 0.01201316,
      "auxiliary_loss_mlp": 0.01038545,
      "balance_loss_clip": 0.93968934,
      "balance_loss_mlp": 1.02901411,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 1.9428497664060231,
      "language_loss": 0.79725564,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.81965423,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.875330924987793
    },
    {
      "auxiliary_loss_clip": 0.0119377,
      "auxiliary_loss_mlp": 0.0103155,
      "balance_loss_clip": 1.01803458,
      "balance_loss_mlp": 1.02200699,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 1.9785685117453184,
      "language_loss": 0.80580401,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.82805717,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.730367422103882
    },
    {
      "auxiliary_loss_clip": 0.01206745,
      "auxiliary_loss_mlp": 0.01031411,
      "balance_loss_clip": 0.98113787,
      "balance_loss_mlp": 1.02186823,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 1.6365166183799285,
      "language_loss": 0.81303692,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.83541846,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 2.696115016937256
    },
    {
      "auxiliary_loss_clip": 0.01201613,
      "auxiliary_loss_mlp": 0.01032064,
      "balance_loss_clip": 0.98296881,
      "balance_loss_mlp": 1.02373147,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 1.7780775213526214,
      "language_loss": 0.80626166,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82859838,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 2.676456928253174
    },
    {
      "auxiliary_loss_clip": 0.01187077,
      "auxiliary_loss_mlp": 0.01028753,
      "balance_loss_clip": 0.82142216,
      "balance_loss_mlp": 1.01950252,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 2.19017024495622,
      "language_loss": 0.70892727,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73108554,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.9399492740631104
    },
    {
      "auxiliary_loss_clip": 0.01201342,
      "auxiliary_loss_mlp": 0.01126281,
      "balance_loss_clip": 1.02136064,
      "balance_loss_mlp": 0.0,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 2.018027086924839,
      "language_loss": 0.80012822,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.82340455,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 3.137984037399292
    },
    {
      "auxiliary_loss_clip": 0.01096972,
      "auxiliary_loss_mlp": 0.01003822,
      "balance_loss_clip": 0.98433983,
      "balance_loss_mlp": 1.00086582,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0031675562425986,
      "language_loss": 0.60326523,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.62427312,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.2322158813476562
    },
    {
      "auxiliary_loss_clip": 0.01205562,
      "auxiliary_loss_mlp": 0.01032117,
      "balance_loss_clip": 1.02097213,
      "balance_loss_mlp": 1.02255678,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 1.9967971341313069,
      "language_loss": 0.778979,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.80135578,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 2.7398428916931152
    },
    {
      "auxiliary_loss_clip": 0.0120217,
      "auxiliary_loss_mlp": 0.01034173,
      "balance_loss_clip": 0.93840027,
      "balance_loss_mlp": 1.02441609,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 2.327100187491823,
      "language_loss": 0.88771963,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.910083,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 3.8159923553466797
    },
    {
      "auxiliary_loss_clip": 0.01200866,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 0.94214571,
      "balance_loss_mlp": 1.0209614,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 1.7594368498036443,
      "language_loss": 0.62845689,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.6507653,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 2.699065923690796
    },
    {
      "auxiliary_loss_clip": 0.01201407,
      "auxiliary_loss_mlp": 0.0103378,
      "balance_loss_clip": 1.05769062,
      "balance_loss_mlp": 1.02455902,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 2.230385509986443,
      "language_loss": 0.85257578,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.8749277,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 3.8513903617858887
    },
    {
      "auxiliary_loss_clip": 0.01201212,
      "auxiliary_loss_mlp": 0.01038081,
      "balance_loss_clip": 1.01982021,
      "balance_loss_mlp": 1.02907443,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 2.0246720609298947,
      "language_loss": 0.88399184,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90638471,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 2.778045892715454
    },
    {
      "auxiliary_loss_clip": 0.01203516,
      "auxiliary_loss_mlp": 0.01040229,
      "balance_loss_clip": 1.02242851,
      "balance_loss_mlp": 1.03082931,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.889456933769567,
      "language_loss": 0.81639171,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.83882916,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.7425220012664795
    },
    {
      "auxiliary_loss_clip": 0.01200836,
      "auxiliary_loss_mlp": 0.01029593,
      "balance_loss_clip": 0.86679411,
      "balance_loss_mlp": 1.0205276,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 1.503934416152367,
      "language_loss": 0.76491988,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.78722417,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 3.96586275100708
    },
    {
      "auxiliary_loss_clip": 0.01202357,
      "auxiliary_loss_mlp": 0.01032315,
      "balance_loss_clip": 1.02013707,
      "balance_loss_mlp": 1.022367,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 1.8728594433772463,
      "language_loss": 0.86329931,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.88564605,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 3.4703354835510254
    },
    {
      "auxiliary_loss_clip": 0.01190139,
      "auxiliary_loss_mlp": 0.01034115,
      "balance_loss_clip": 0.94093513,
      "balance_loss_mlp": 1.0254066,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.8487965823988575,
      "language_loss": 0.84181428,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86405683,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 2.6593286991119385
    },
    {
      "auxiliary_loss_clip": 0.01090417,
      "auxiliary_loss_mlp": 0.01007809,
      "balance_loss_clip": 0.94427252,
      "balance_loss_mlp": 1.00473309,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.903727202631929,
      "language_loss": 0.67899871,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.69998097,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.293757200241089
    },
    {
      "auxiliary_loss_clip": 0.01198613,
      "auxiliary_loss_mlp": 0.01036154,
      "balance_loss_clip": 0.98201621,
      "balance_loss_mlp": 1.0277493,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 2.201870393733751,
      "language_loss": 0.86134672,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.88369435,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 2.6955909729003906
    },
    {
      "auxiliary_loss_clip": 0.01198243,
      "auxiliary_loss_mlp": 0.01030814,
      "balance_loss_clip": 0.94451177,
      "balance_loss_mlp": 1.02109277,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 1.777485230454883,
      "language_loss": 0.84928393,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.87157446,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.7368342876434326
    },
    {
      "auxiliary_loss_clip": 0.01200909,
      "auxiliary_loss_mlp": 0.01035827,
      "balance_loss_clip": 1.02216625,
      "balance_loss_mlp": 1.02701116,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 1.9639037864391948,
      "language_loss": 0.66421354,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.6865809,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.736133337020874
    },
    {
      "auxiliary_loss_clip": 0.01202123,
      "auxiliary_loss_mlp": 0.01049691,
      "balance_loss_clip": 1.02197254,
      "balance_loss_mlp": 1.03973091,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 2.1080205978190394,
      "language_loss": 0.81826079,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84077895,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.8068342208862305
    },
    {
      "auxiliary_loss_clip": 0.01202065,
      "auxiliary_loss_mlp": 0.01032663,
      "balance_loss_clip": 1.05860281,
      "balance_loss_mlp": 1.02402639,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 2.481595914335354,
      "language_loss": 0.69052368,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.71287102,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.682800531387329
    },
    {
      "auxiliary_loss_clip": 0.01205669,
      "auxiliary_loss_mlp": 0.01034084,
      "balance_loss_clip": 0.94289839,
      "balance_loss_mlp": 1.02408838,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 2.3077656519850724,
      "language_loss": 0.79612857,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.81852615,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 2.728415012359619
    },
    {
      "auxiliary_loss_clip": 0.01201232,
      "auxiliary_loss_mlp": 0.01028467,
      "balance_loss_clip": 0.94292361,
      "balance_loss_mlp": 1.0194726,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 1.7341341036939817,
      "language_loss": 0.67562431,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.69792128,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.6496734619140625
    },
    {
      "auxiliary_loss_clip": 0.01189732,
      "auxiliary_loss_mlp": 0.01030575,
      "balance_loss_clip": 1.01734579,
      "balance_loss_mlp": 1.02076983,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 1.9409369400225738,
      "language_loss": 0.72666728,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.74887031,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.6447713375091553
    },
    {
      "auxiliary_loss_clip": 0.01201683,
      "auxiliary_loss_mlp": 0.01029338,
      "balance_loss_clip": 0.98053408,
      "balance_loss_mlp": 1.01938987,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 1.8513338048901606,
      "language_loss": 0.8532908,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.87560105,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.684168815612793
    },
    {
      "auxiliary_loss_clip": 0.01105909,
      "auxiliary_loss_mlp": 0.01004383,
      "balance_loss_clip": 0.87787747,
      "balance_loss_mlp": 1.00102139,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.7285652910090866,
      "language_loss": 0.55987179,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58097476,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.3153493404388428
    },
    {
      "auxiliary_loss_clip": 0.01179412,
      "auxiliary_loss_mlp": 0.01038621,
      "balance_loss_clip": 0.89955598,
      "balance_loss_mlp": 1.02894664,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 1.9887050759693945,
      "language_loss": 0.77646279,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.79864311,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.936155319213867
    },
    {
      "auxiliary_loss_clip": 0.0120276,
      "auxiliary_loss_mlp": 0.01030795,
      "balance_loss_clip": 0.98284888,
      "balance_loss_mlp": 1.02168727,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 2.2737932809834636,
      "language_loss": 0.79274201,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.81507754,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 2.688434362411499
    },
    {
      "auxiliary_loss_clip": 0.01207471,
      "auxiliary_loss_mlp": 0.01126288,
      "balance_loss_clip": 1.02040815,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 2.2534014095874024,
      "language_loss": 0.78800666,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.81134427,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 2.754831552505493
    },
    {
      "auxiliary_loss_clip": 0.01200037,
      "auxiliary_loss_mlp": 0.01031688,
      "balance_loss_clip": 0.9447881,
      "balance_loss_mlp": 1.02246165,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 2.0942555824080182,
      "language_loss": 0.84890711,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.87122428,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.7823634147644043
    },
    {
      "auxiliary_loss_clip": 0.01202346,
      "auxiliary_loss_mlp": 0.01033854,
      "balance_loss_clip": 0.98144472,
      "balance_loss_mlp": 1.02484822,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 3.0479852718182503,
      "language_loss": 0.64803386,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67039585,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.884507894515991
    },
    {
      "auxiliary_loss_clip": 0.01185103,
      "auxiliary_loss_mlp": 0.01034425,
      "balance_loss_clip": 0.94343776,
      "balance_loss_mlp": 1.02520418,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 2.0643115995627697,
      "language_loss": 0.82558215,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84777749,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.784229278564453
    },
    {
      "auxiliary_loss_clip": 0.01202778,
      "auxiliary_loss_mlp": 0.01038938,
      "balance_loss_clip": 0.98418808,
      "balance_loss_mlp": 1.02991998,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 1.574941683767417,
      "language_loss": 0.79004169,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.81245887,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 3.7397711277008057
    },
    {
      "auxiliary_loss_clip": 0.01185888,
      "auxiliary_loss_mlp": 0.0103123,
      "balance_loss_clip": 0.94139153,
      "balance_loss_mlp": 1.02226543,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 2.3735329455128644,
      "language_loss": 0.68782675,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70999789,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.7413711547851562
    },
    {
      "auxiliary_loss_clip": 0.01196628,
      "auxiliary_loss_mlp": 0.01037015,
      "balance_loss_clip": 0.98117596,
      "balance_loss_mlp": 1.02837217,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 1.7122735204189754,
      "language_loss": 0.73233312,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.75466955,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 2.699871778488159
    },
    {
      "auxiliary_loss_clip": 0.01199433,
      "auxiliary_loss_mlp": 0.01033459,
      "balance_loss_clip": 1.02181125,
      "balance_loss_mlp": 1.02416039,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 2.1775419711571518,
      "language_loss": 0.75165921,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77398819,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 2.7435853481292725
    },
    {
      "auxiliary_loss_clip": 0.01202476,
      "auxiliary_loss_mlp": 0.01034913,
      "balance_loss_clip": 0.98227227,
      "balance_loss_mlp": 1.02628803,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 2.466974694583679,
      "language_loss": 0.70510042,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.72747427,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 3.820847749710083
    },
    {
      "auxiliary_loss_clip": 0.01203952,
      "auxiliary_loss_mlp": 0.01125943,
      "balance_loss_clip": 1.01804233,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 1.8016935677675059,
      "language_loss": 0.70138526,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72468424,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.678868293762207
    },
    {
      "auxiliary_loss_clip": 0.01189359,
      "auxiliary_loss_mlp": 0.01036213,
      "balance_loss_clip": 0.86990631,
      "balance_loss_mlp": 1.02752256,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 3.220208408274187,
      "language_loss": 0.72837782,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.7506336,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 3.8990540504455566
    },
    {
      "auxiliary_loss_clip": 0.01205318,
      "auxiliary_loss_mlp": 0.01030809,
      "balance_loss_clip": 0.94368571,
      "balance_loss_mlp": 1.02109933,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.4734518299111135,
      "language_loss": 0.77664697,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.79900819,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 3.640841245651245
    },
    {
      "auxiliary_loss_clip": 0.01206787,
      "auxiliary_loss_mlp": 0.01125875,
      "balance_loss_clip": 1.06017876,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 2.1220830331470175,
      "language_loss": 0.73212671,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.75545335,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.707937240600586
    },
    {
      "auxiliary_loss_clip": 0.01206759,
      "auxiliary_loss_mlp": 0.01125308,
      "balance_loss_clip": 1.02409244,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 1.7497199819854754,
      "language_loss": 0.81558919,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83890986,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 2.7965846061706543
    },
    {
      "auxiliary_loss_clip": 0.01201808,
      "auxiliary_loss_mlp": 0.01033215,
      "balance_loss_clip": 0.98454499,
      "balance_loss_mlp": 1.02379751,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 1.71655433160864,
      "language_loss": 0.89487934,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91722953,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 2.6909642219543457
    },
    {
      "auxiliary_loss_clip": 0.01205255,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.06148994,
      "balance_loss_mlp": 1.02198648,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 1.8115312894128641,
      "language_loss": 0.66948485,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.6918484,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.8525869846343994
    },
    {
      "auxiliary_loss_clip": 0.01192874,
      "auxiliary_loss_mlp": 0.01033307,
      "balance_loss_clip": 0.94253665,
      "balance_loss_mlp": 1.02421117,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 2.1950383873453916,
      "language_loss": 0.88405865,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90632045,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 2.7819738388061523
    },
    {
      "auxiliary_loss_clip": 0.01195372,
      "auxiliary_loss_mlp": 0.0103129,
      "balance_loss_clip": 0.86339658,
      "balance_loss_mlp": 1.02126408,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.247549233992048,
      "language_loss": 0.6422379,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66450453,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.869983196258545
    },
    {
      "auxiliary_loss_clip": 0.01194344,
      "auxiliary_loss_mlp": 0.01033196,
      "balance_loss_clip": 0.98327458,
      "balance_loss_mlp": 1.02423787,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 1.859930713198558,
      "language_loss": 0.63866723,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66094261,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.796905279159546
    },
    {
      "auxiliary_loss_clip": 0.01190292,
      "auxiliary_loss_mlp": 0.01037347,
      "balance_loss_clip": 0.90245891,
      "balance_loss_mlp": 1.02790594,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 1.7312871061377877,
      "language_loss": 0.68548369,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.7077601,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.9326674938201904
    },
    {
      "auxiliary_loss_clip": 0.01189626,
      "auxiliary_loss_mlp": 0.01034684,
      "balance_loss_clip": 0.94361782,
      "balance_loss_mlp": 1.02589834,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.34863905846847,
      "language_loss": 0.85116303,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87340611,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 2.7818195819854736
    },
    {
      "auxiliary_loss_clip": 0.01207495,
      "auxiliary_loss_mlp": 0.01044936,
      "balance_loss_clip": 1.06107712,
      "balance_loss_mlp": 1.03547049,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 2.040246926914697,
      "language_loss": 0.72036082,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.74288511,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 2.6605374813079834
    },
    {
      "auxiliary_loss_clip": 0.01201802,
      "auxiliary_loss_mlp": 0.01030651,
      "balance_loss_clip": 1.02071238,
      "balance_loss_mlp": 1.02057219,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 2.073837195105151,
      "language_loss": 0.77436364,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.7966882,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.7456867694854736
    },
    {
      "auxiliary_loss_clip": 0.01163707,
      "auxiliary_loss_mlp": 0.01126485,
      "balance_loss_clip": 0.81932378,
      "balance_loss_mlp": 0.0,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 1.8896566280984528,
      "language_loss": 0.80040598,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.82330787,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.8743367195129395
    },
    {
      "auxiliary_loss_clip": 0.01103317,
      "auxiliary_loss_mlp": 0.0101665,
      "balance_loss_clip": 0.95113748,
      "balance_loss_mlp": 1.01377714,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.8614690619440338,
      "language_loss": 0.61253589,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.63373554,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.665418863296509
    },
    {
      "auxiliary_loss_clip": 0.01203006,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.02300811,
      "balance_loss_mlp": 1.02397883,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 1.8574983769015148,
      "language_loss": 0.76762569,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.78999102,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 2.7899703979492188
    },
    {
      "auxiliary_loss_clip": 0.0120156,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 0.94362998,
      "balance_loss_mlp": 1.02232111,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 2.1209514470410538,
      "language_loss": 0.77556515,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.79789662,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 3.1277096271514893
    },
    {
      "auxiliary_loss_clip": 0.01208627,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 1.02271068,
      "balance_loss_mlp": 1.02148485,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 1.8619188802828006,
      "language_loss": 0.87776619,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90016639,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.7451171875
    },
    {
      "auxiliary_loss_clip": 0.01201998,
      "auxiliary_loss_mlp": 0.01036305,
      "balance_loss_clip": 1.02322304,
      "balance_loss_mlp": 1.02737665,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 2.120015354603101,
      "language_loss": 0.88423288,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90661585,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.6420230865478516
    },
    {
      "auxiliary_loss_clip": 0.01198081,
      "auxiliary_loss_mlp": 0.010343,
      "balance_loss_clip": 1.02044821,
      "balance_loss_mlp": 1.02507341,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 1.8625766536943218,
      "language_loss": 0.77875543,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.80107927,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.677917003631592
    },
    {
      "auxiliary_loss_clip": 0.01092006,
      "auxiliary_loss_mlp": 0.01008137,
      "balance_loss_clip": 0.98480064,
      "balance_loss_mlp": 1.00525224,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.7880549536188021,
      "language_loss": 0.59718138,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61818278,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 4.477405548095703
    },
    {
      "auxiliary_loss_clip": 0.01197191,
      "auxiliary_loss_mlp": 0.01041192,
      "balance_loss_clip": 0.9087317,
      "balance_loss_mlp": 1.03210247,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 1.8417112176132298,
      "language_loss": 0.76424766,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.78663152,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.799269914627075
    },
    {
      "auxiliary_loss_clip": 0.01207488,
      "auxiliary_loss_mlp": 0.01031559,
      "balance_loss_clip": 0.9847278,
      "balance_loss_mlp": 1.02266598,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 1.6788588881106112,
      "language_loss": 0.80842358,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.830814,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.7238893508911133
    },
    {
      "auxiliary_loss_clip": 0.01194049,
      "auxiliary_loss_mlp": 0.01034868,
      "balance_loss_clip": 1.01958251,
      "balance_loss_mlp": 1.02599239,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 1.7799450789387796,
      "language_loss": 0.76993585,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.792225,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 3.7109577655792236
    },
    {
      "auxiliary_loss_clip": 0.01194263,
      "auxiliary_loss_mlp": 0.01037263,
      "balance_loss_clip": 0.97823477,
      "balance_loss_mlp": 1.02878702,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 2.008621665356517,
      "language_loss": 0.77370518,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.79602045,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 2.7655258178710938
    },
    {
      "auxiliary_loss_clip": 0.01169481,
      "auxiliary_loss_mlp": 0.01029274,
      "balance_loss_clip": 0.86329615,
      "balance_loss_mlp": 1.020715,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 1.8282945977925398,
      "language_loss": 0.77953595,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.80152351,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.8433711528778076
    },
    {
      "auxiliary_loss_clip": 0.01202636,
      "auxiliary_loss_mlp": 0.01037586,
      "balance_loss_clip": 1.06064618,
      "balance_loss_mlp": 1.02918744,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 1.6475391816449279,
      "language_loss": 0.79336846,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.81577069,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.6849985122680664
    },
    {
      "auxiliary_loss_clip": 0.01208662,
      "auxiliary_loss_mlp": 0.0104373,
      "balance_loss_clip": 0.94843721,
      "balance_loss_mlp": 1.03419328,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 4.366687555616971,
      "language_loss": 0.86883491,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.89135879,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 3.6348865032196045
    },
    {
      "auxiliary_loss_clip": 0.01200318,
      "auxiliary_loss_mlp": 0.01037235,
      "balance_loss_clip": 0.86817968,
      "balance_loss_mlp": 1.02857399,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.7185449194200753,
      "language_loss": 0.78626579,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.80864131,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.7967872619628906
    },
    {
      "auxiliary_loss_clip": 0.01088757,
      "auxiliary_loss_mlp": 0.01120699,
      "balance_loss_clip": 0.94551504,
      "balance_loss_mlp": 0.0,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.6818989561709633,
      "language_loss": 0.54306567,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56516021,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.3285820484161377
    },
    {
      "auxiliary_loss_clip": 0.01195693,
      "auxiliary_loss_mlp": 0.01035052,
      "balance_loss_clip": 0.94402885,
      "balance_loss_mlp": 1.02580702,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 2.0550035676429923,
      "language_loss": 0.89191341,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91422093,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.8098878860473633
    },
    {
      "auxiliary_loss_clip": 0.01205367,
      "auxiliary_loss_mlp": 0.01030139,
      "balance_loss_clip": 0.98405111,
      "balance_loss_mlp": 1.02174652,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 1.935030441223049,
      "language_loss": 0.75913537,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.78149045,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 2.810328722000122
    },
    {
      "auxiliary_loss_clip": 0.01199089,
      "auxiliary_loss_mlp": 0.01033575,
      "balance_loss_clip": 0.86701107,
      "balance_loss_mlp": 1.02298284,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 1.8571487563919657,
      "language_loss": 0.6623081,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.68463469,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.8498120307922363
    },
    {
      "auxiliary_loss_clip": 0.01089588,
      "auxiliary_loss_mlp": 0.01004007,
      "balance_loss_clip": 1.01973808,
      "balance_loss_mlp": 1.001122,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.8826277316212152,
      "language_loss": 0.60714608,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62808198,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 3.0844039916992188
    },
    {
      "auxiliary_loss_clip": 0.01199805,
      "auxiliary_loss_mlp": 0.01044721,
      "balance_loss_clip": 0.98098713,
      "balance_loss_mlp": 1.03588748,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 2.3799216428694447,
      "language_loss": 0.91487384,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.93731916,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.724062204360962
    },
    {
      "auxiliary_loss_clip": 0.01197926,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.05670762,
      "balance_loss_mlp": 1.01743889,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.488374718308427,
      "language_loss": 0.81621879,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.83846158,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.7213189601898193
    },
    {
      "auxiliary_loss_clip": 0.01199674,
      "auxiliary_loss_mlp": 0.01027654,
      "balance_loss_clip": 0.98191142,
      "balance_loss_mlp": 1.0188266,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 1.575782011488699,
      "language_loss": 0.8721993,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.8944726,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 2.6547138690948486
    },
    {
      "auxiliary_loss_clip": 0.0120181,
      "auxiliary_loss_mlp": 0.01039996,
      "balance_loss_clip": 1.05818343,
      "balance_loss_mlp": 1.02979159,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 2.1961457323481226,
      "language_loss": 0.82698882,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.84940684,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 2.67177677154541
    },
    {
      "auxiliary_loss_clip": 0.01196204,
      "auxiliary_loss_mlp": 0.01030982,
      "balance_loss_clip": 0.94037008,
      "balance_loss_mlp": 1.02282166,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 3.159848511694097,
      "language_loss": 0.80934417,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.83161604,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.77787446975708
    },
    {
      "auxiliary_loss_clip": 0.01084094,
      "auxiliary_loss_mlp": 0.01002931,
      "balance_loss_clip": 0.97875535,
      "balance_loss_mlp": 0.99983138,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7583358183491319,
      "language_loss": 0.51853579,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.53940606,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.3835220336914062
    },
    {
      "auxiliary_loss_clip": 0.01202565,
      "auxiliary_loss_mlp": 0.01039509,
      "balance_loss_clip": 0.90619886,
      "balance_loss_mlp": 1.02971029,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 2.2566146738923645,
      "language_loss": 0.79597068,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.81839144,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.8408665657043457
    },
    {
      "auxiliary_loss_clip": 0.01093675,
      "auxiliary_loss_mlp": 0.01003089,
      "balance_loss_clip": 0.94512665,
      "balance_loss_mlp": 1.0001086,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8264307667147652,
      "language_loss": 0.62559074,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64655834,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 3.3496391773223877
    },
    {
      "auxiliary_loss_clip": 0.01199987,
      "auxiliary_loss_mlp": 0.01037062,
      "balance_loss_clip": 1.02083611,
      "balance_loss_mlp": 1.02758455,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 2.370918851016458,
      "language_loss": 0.63617659,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.65854704,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 2.7631657123565674
    },
    {
      "auxiliary_loss_clip": 0.01197262,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 0.98188281,
      "balance_loss_mlp": 1.01651919,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 1.8508804461743922,
      "language_loss": 0.75653291,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.77875865,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.7924644947052
    },
    {
      "auxiliary_loss_clip": 0.01198834,
      "auxiliary_loss_mlp": 0.01029831,
      "balance_loss_clip": 1.01979232,
      "balance_loss_mlp": 1.02187419,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 1.7993305855932253,
      "language_loss": 0.82220763,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84449428,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.8368849754333496
    },
    {
      "auxiliary_loss_clip": 0.01093297,
      "auxiliary_loss_mlp": 0.01120638,
      "balance_loss_clip": 0.9058888,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8775179114805267,
      "language_loss": 0.63749731,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.65963668,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.3587701320648193
    },
    {
      "auxiliary_loss_clip": 0.01190312,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 0.94046247,
      "balance_loss_mlp": 1.02013779,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 2.0563999039336998,
      "language_loss": 0.80852342,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.83071375,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 3.8780314922332764
    },
    {
      "auxiliary_loss_clip": 0.01094049,
      "auxiliary_loss_mlp": 0.01005781,
      "balance_loss_clip": 0.94365668,
      "balance_loss_mlp": 1.00293207,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.8144689357071289,
      "language_loss": 0.60472643,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62572467,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.269415855407715
    },
    {
      "auxiliary_loss_clip": 0.0119766,
      "auxiliary_loss_mlp": 0.01030882,
      "balance_loss_clip": 1.05664456,
      "balance_loss_mlp": 1.02236462,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 1.6523833163470767,
      "language_loss": 0.79166663,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81395209,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.6928746700286865
    },
    {
      "auxiliary_loss_clip": 0.01200322,
      "auxiliary_loss_mlp": 0.01035625,
      "balance_loss_clip": 0.90253854,
      "balance_loss_mlp": 1.0261538,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 2.099696673318503,
      "language_loss": 0.63388324,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65624273,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 3.8674163818359375
    },
    {
      "auxiliary_loss_clip": 0.01200426,
      "auxiliary_loss_mlp": 0.01038985,
      "balance_loss_clip": 1.02270722,
      "balance_loss_mlp": 1.03033042,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 1.7729458924205288,
      "language_loss": 0.7750476,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.79744172,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 2.7170534133911133
    },
    {
      "auxiliary_loss_clip": 0.01200484,
      "auxiliary_loss_mlp": 0.01039314,
      "balance_loss_clip": 0.90531665,
      "balance_loss_mlp": 1.02969348,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 2.584998289120869,
      "language_loss": 0.78736216,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.80976009,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.689401865005493
    },
    {
      "auxiliary_loss_clip": 0.01188973,
      "auxiliary_loss_mlp": 0.01034226,
      "balance_loss_clip": 0.90403742,
      "balance_loss_mlp": 1.0252676,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 1.6953506098381024,
      "language_loss": 0.78145057,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.80368257,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.8256256580352783
    },
    {
      "auxiliary_loss_clip": 0.01203626,
      "auxiliary_loss_mlp": 0.01032449,
      "balance_loss_clip": 1.05737591,
      "balance_loss_mlp": 1.02322769,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 2.1556237419120774,
      "language_loss": 0.83396816,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85632896,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 5.079745531082153
    },
    {
      "auxiliary_loss_clip": 0.01186848,
      "auxiliary_loss_mlp": 0.01125618,
      "balance_loss_clip": 0.94240153,
      "balance_loss_mlp": 0.0,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 3.1268790334952734,
      "language_loss": 0.75303757,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77616227,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.920599937438965
    },
    {
      "auxiliary_loss_clip": 0.01197812,
      "auxiliary_loss_mlp": 0.01037128,
      "balance_loss_clip": 1.01943517,
      "balance_loss_mlp": 1.02768028,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 1.9843053786698657,
      "language_loss": 0.84594959,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.86829901,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.6330037117004395
    },
    {
      "auxiliary_loss_clip": 0.01198699,
      "auxiliary_loss_mlp": 0.01028113,
      "balance_loss_clip": 1.02005744,
      "balance_loss_mlp": 1.0190289,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 2.5828176888581913,
      "language_loss": 0.66075075,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.68301892,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.6052918434143066
    },
    {
      "auxiliary_loss_clip": 0.01189581,
      "auxiliary_loss_mlp": 0.01023508,
      "balance_loss_clip": 1.01778412,
      "balance_loss_mlp": 1.01465058,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 1.8177780647057802,
      "language_loss": 0.80540693,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82753778,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 2.6552140712738037
    },
    {
      "auxiliary_loss_clip": 0.01087541,
      "auxiliary_loss_mlp": 0.01006959,
      "balance_loss_clip": 1.01799834,
      "balance_loss_mlp": 1.00419331,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7836613297682147,
      "language_loss": 0.55990899,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.58085406,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.198430299758911
    },
    {
      "auxiliary_loss_clip": 0.01195303,
      "auxiliary_loss_mlp": 0.01031621,
      "balance_loss_clip": 0.97791243,
      "balance_loss_mlp": 1.02285933,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 1.643831767967308,
      "language_loss": 0.80692941,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.82919866,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.7509541511535645
    },
    {
      "auxiliary_loss_clip": 0.01199513,
      "auxiliary_loss_mlp": 0.01035126,
      "balance_loss_clip": 1.05560553,
      "balance_loss_mlp": 1.02590513,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 2.008680948641446,
      "language_loss": 0.78060305,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80294949,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.6323883533477783
    },
    {
      "auxiliary_loss_clip": 0.01193631,
      "auxiliary_loss_mlp": 0.01039363,
      "balance_loss_clip": 1.01886868,
      "balance_loss_mlp": 1.03093505,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 1.8730972980649114,
      "language_loss": 0.87618113,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89851105,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.6918678283691406
    },
    {
      "auxiliary_loss_clip": 0.01187308,
      "auxiliary_loss_mlp": 0.01033891,
      "balance_loss_clip": 0.93809992,
      "balance_loss_mlp": 1.02461684,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 2.060883276419869,
      "language_loss": 0.9047879,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92699987,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 2.773925542831421
    },
    {
      "auxiliary_loss_clip": 0.01196795,
      "auxiliary_loss_mlp": 0.01029143,
      "balance_loss_clip": 0.97826064,
      "balance_loss_mlp": 1.02013123,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 1.8964648697326467,
      "language_loss": 0.78127587,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80353522,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 2.689737319946289
    },
    {
      "auxiliary_loss_clip": 0.01202315,
      "auxiliary_loss_mlp": 0.01125276,
      "balance_loss_clip": 0.97841406,
      "balance_loss_mlp": 0.0,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.553255394957389,
      "language_loss": 0.80129182,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.82456779,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.7300171852111816
    },
    {
      "auxiliary_loss_clip": 0.01196006,
      "auxiliary_loss_mlp": 0.01034467,
      "balance_loss_clip": 1.01902699,
      "balance_loss_mlp": 1.02539515,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 1.7274544898117525,
      "language_loss": 0.72279215,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.7450968,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 2.690270185470581
    },
    {
      "auxiliary_loss_clip": 0.01203151,
      "auxiliary_loss_mlp": 0.01125427,
      "balance_loss_clip": 0.98169452,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 1.5086789985316011,
      "language_loss": 0.72142076,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.74470651,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.7653791904449463
    },
    {
      "auxiliary_loss_clip": 0.01185669,
      "auxiliary_loss_mlp": 0.01039451,
      "balance_loss_clip": 0.90382421,
      "balance_loss_mlp": 1.03021812,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.032405368762612,
      "language_loss": 0.84664178,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.86889303,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 2.7615163326263428
    },
    {
      "auxiliary_loss_clip": 0.01199334,
      "auxiliary_loss_mlp": 0.01125454,
      "balance_loss_clip": 1.05707622,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 1.9778987605420246,
      "language_loss": 0.65446889,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.67771673,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.6956615447998047
    },
    {
      "auxiliary_loss_clip": 0.01199361,
      "auxiliary_loss_mlp": 0.01030833,
      "balance_loss_clip": 0.98148066,
      "balance_loss_mlp": 1.02213717,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.109860494158819,
      "language_loss": 0.8321991,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.85450101,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.745185613632202
    },
    {
      "auxiliary_loss_clip": 0.01196018,
      "auxiliary_loss_mlp": 0.0103551,
      "balance_loss_clip": 0.98111105,
      "balance_loss_mlp": 1.02711141,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 1.6637417687388494,
      "language_loss": 0.7058534,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.72816867,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.7310004234313965
    },
    {
      "auxiliary_loss_clip": 0.01196441,
      "auxiliary_loss_mlp": 0.01031635,
      "balance_loss_clip": 0.94153738,
      "balance_loss_mlp": 1.02238989,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 2.13491841480153,
      "language_loss": 0.73852056,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.76080132,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.7467784881591797
    },
    {
      "auxiliary_loss_clip": 0.01194649,
      "auxiliary_loss_mlp": 0.01036904,
      "balance_loss_clip": 1.01674414,
      "balance_loss_mlp": 1.0286901,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.7808018930853966,
      "language_loss": 0.87396967,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89628518,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 3.646397113800049
    },
    {
      "auxiliary_loss_clip": 0.01202793,
      "auxiliary_loss_mlp": 0.01031508,
      "balance_loss_clip": 1.05868495,
      "balance_loss_mlp": 1.02243638,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.5837946126567262,
      "language_loss": 0.81943733,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.8417803,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 2.6230483055114746
    },
    {
      "auxiliary_loss_clip": 0.0119672,
      "auxiliary_loss_mlp": 0.01027288,
      "balance_loss_clip": 0.9021796,
      "balance_loss_mlp": 1.01854968,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.095116758845134,
      "language_loss": 0.81924212,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84148228,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 2.728867769241333
    },
    {
      "auxiliary_loss_clip": 0.01191621,
      "auxiliary_loss_mlp": 0.0103824,
      "balance_loss_clip": 1.01886225,
      "balance_loss_mlp": 1.0299077,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 3.022430103254836,
      "language_loss": 0.76024395,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.78254259,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 3.6939854621887207
    },
    {
      "auxiliary_loss_clip": 0.01196663,
      "auxiliary_loss_mlp": 0.01036911,
      "balance_loss_clip": 1.01903915,
      "balance_loss_mlp": 1.02854812,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 2.3732172979784796,
      "language_loss": 0.76624978,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.78858554,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.6758830547332764
    },
    {
      "auxiliary_loss_clip": 0.01195788,
      "auxiliary_loss_mlp": 0.01035703,
      "balance_loss_clip": 0.97809827,
      "balance_loss_mlp": 1.02696538,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 1.4760882940942166,
      "language_loss": 0.77576512,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.79808009,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 2.775662660598755
    },
    {
      "auxiliary_loss_clip": 0.01208932,
      "auxiliary_loss_mlp": 0.01035162,
      "balance_loss_clip": 0.98376566,
      "balance_loss_mlp": 1.02614355,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 2.0487287976390074,
      "language_loss": 0.7591573,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78159833,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 3.638982057571411
    },
    {
      "auxiliary_loss_clip": 0.01200068,
      "auxiliary_loss_mlp": 0.0103326,
      "balance_loss_clip": 0.94437081,
      "balance_loss_mlp": 1.0231626,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 3.3588385952861857,
      "language_loss": 0.80006945,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82240278,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 3.5522475242614746
    },
    {
      "auxiliary_loss_clip": 0.01195506,
      "auxiliary_loss_mlp": 0.01032614,
      "balance_loss_clip": 1.01719093,
      "balance_loss_mlp": 1.02347088,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 2.2724577979611866,
      "language_loss": 0.81455475,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83683592,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.611419439315796
    },
    {
      "auxiliary_loss_clip": 0.01188742,
      "auxiliary_loss_mlp": 0.01125587,
      "balance_loss_clip": 0.9784416,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 1.9124676154456504,
      "language_loss": 0.67571199,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.69885528,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.7056756019592285
    },
    {
      "auxiliary_loss_clip": 0.01190871,
      "auxiliary_loss_mlp": 0.01036313,
      "balance_loss_clip": 1.01722908,
      "balance_loss_mlp": 1.02713394,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 2.2411522471879053,
      "language_loss": 0.75313401,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77540588,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 2.6700000762939453
    },
    {
      "auxiliary_loss_clip": 0.0119679,
      "auxiliary_loss_mlp": 0.01031697,
      "balance_loss_clip": 0.97791135,
      "balance_loss_mlp": 1.02276206,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.2127716347319177,
      "language_loss": 0.79342401,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.81570888,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.660386562347412
    },
    {
      "auxiliary_loss_clip": 0.01194169,
      "auxiliary_loss_mlp": 0.01033765,
      "balance_loss_clip": 1.01505566,
      "balance_loss_mlp": 1.0253365,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 2.5436699901502324,
      "language_loss": 0.78523719,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80751652,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 2.653716802597046
    },
    {
      "auxiliary_loss_clip": 0.01196274,
      "auxiliary_loss_mlp": 0.01036073,
      "balance_loss_clip": 0.98167604,
      "balance_loss_mlp": 1.02702451,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 1.9497840690611346,
      "language_loss": 0.73846751,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76079094,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.768099069595337
    },
    {
      "auxiliary_loss_clip": 0.01181391,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 0.90175581,
      "balance_loss_mlp": 1.01771688,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 1.9298319510118285,
      "language_loss": 0.77724016,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.79932201,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.7745518684387207
    },
    {
      "auxiliary_loss_clip": 0.01197937,
      "auxiliary_loss_mlp": 0.01031038,
      "balance_loss_clip": 1.05493283,
      "balance_loss_mlp": 1.02184701,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 1.7413902213022643,
      "language_loss": 0.82454681,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84683657,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.714779853820801
    },
    {
      "auxiliary_loss_clip": 0.0119529,
      "auxiliary_loss_mlp": 0.0112568,
      "balance_loss_clip": 0.9791435,
      "balance_loss_mlp": 0.0,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.6757478817663007,
      "language_loss": 0.81512415,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.83833385,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.7161710262298584
    },
    {
      "auxiliary_loss_clip": 0.01198672,
      "auxiliary_loss_mlp": 0.0102577,
      "balance_loss_clip": 0.94047987,
      "balance_loss_mlp": 1.01687145,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.4088489847989942,
      "language_loss": 0.77243543,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.79467982,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.7446823120117188
    },
    {
      "auxiliary_loss_clip": 0.01200162,
      "auxiliary_loss_mlp": 0.01125202,
      "balance_loss_clip": 0.94327235,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 2.306785504506554,
      "language_loss": 0.85349894,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.87675256,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 2.664123058319092
    },
    {
      "auxiliary_loss_clip": 0.01202117,
      "auxiliary_loss_mlp": 0.01032192,
      "balance_loss_clip": 1.05836964,
      "balance_loss_mlp": 1.02350712,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 1.695463508407054,
      "language_loss": 0.78665954,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80900264,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 2.627285957336426
    },
    {
      "auxiliary_loss_clip": 0.01191102,
      "auxiliary_loss_mlp": 0.01125388,
      "balance_loss_clip": 0.97827983,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 2.0712214380281733,
      "language_loss": 0.69475096,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71791583,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.6978375911712646
    },
    {
      "auxiliary_loss_clip": 0.01192506,
      "auxiliary_loss_mlp": 0.01035374,
      "balance_loss_clip": 0.90006804,
      "balance_loss_mlp": 1.02670169,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 1.923792727630622,
      "language_loss": 0.60680521,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.62908399,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 2.79121732711792
    },
    {
      "auxiliary_loss_clip": 0.01198256,
      "auxiliary_loss_mlp": 0.01031318,
      "balance_loss_clip": 0.90262473,
      "balance_loss_mlp": 1.02221584,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 1.880895176354405,
      "language_loss": 0.76033849,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.7826342,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.815709352493286
    },
    {
      "auxiliary_loss_clip": 0.01194446,
      "auxiliary_loss_mlp": 0.01125237,
      "balance_loss_clip": 0.97888607,
      "balance_loss_mlp": 0.0,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.162797629038878,
      "language_loss": 0.80073214,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.82392901,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.763310670852661
    },
    {
      "auxiliary_loss_clip": 0.0109802,
      "auxiliary_loss_mlp": 0.0099984,
      "balance_loss_clip": 0.90759331,
      "balance_loss_mlp": 0.9972412,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.8961493335963514,
      "language_loss": 0.70595104,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72692961,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.347461223602295
    },
    {
      "auxiliary_loss_clip": 0.01198941,
      "auxiliary_loss_mlp": 0.01031989,
      "balance_loss_clip": 0.97952771,
      "balance_loss_mlp": 1.02328038,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 1.5792377981648578,
      "language_loss": 0.77920431,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.80151361,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.82491135597229
    },
    {
      "auxiliary_loss_clip": 0.01196292,
      "auxiliary_loss_mlp": 0.01033452,
      "balance_loss_clip": 1.01808643,
      "balance_loss_mlp": 1.02423096,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 2.1076071293907885,
      "language_loss": 0.82041097,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84270835,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 3.809936285018921
    },
    {
      "auxiliary_loss_clip": 0.01202297,
      "auxiliary_loss_mlp": 0.01031535,
      "balance_loss_clip": 1.02035213,
      "balance_loss_mlp": 1.02316666,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 2.1226187399110485,
      "language_loss": 0.82722688,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.84956521,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 2.787773370742798
    },
    {
      "auxiliary_loss_clip": 0.01189682,
      "auxiliary_loss_mlp": 0.01039022,
      "balance_loss_clip": 0.93868148,
      "balance_loss_mlp": 1.02973545,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 1.8488296921168805,
      "language_loss": 0.74222082,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.76450783,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 3.7990808486938477
    },
    {
      "auxiliary_loss_clip": 0.01205148,
      "auxiliary_loss_mlp": 0.01034157,
      "balance_loss_clip": 0.983899,
      "balance_loss_mlp": 1.02575815,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.150090295507856,
      "language_loss": 0.75510526,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.7774983,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 2.8107502460479736
    },
    {
      "auxiliary_loss_clip": 0.01204692,
      "auxiliary_loss_mlp": 0.01032262,
      "balance_loss_clip": 1.06071401,
      "balance_loss_mlp": 1.02363122,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 2.2299184578558453,
      "language_loss": 0.8075369,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.8299064,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 2.751105785369873
    },
    {
      "auxiliary_loss_clip": 0.01203246,
      "auxiliary_loss_mlp": 0.01030923,
      "balance_loss_clip": 1.05998969,
      "balance_loss_mlp": 1.02118349,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 2.33284539798361,
      "language_loss": 0.87561834,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.89796007,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.648090124130249
    },
    {
      "auxiliary_loss_clip": 0.01200628,
      "auxiliary_loss_mlp": 0.0112613,
      "balance_loss_clip": 1.02140975,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 2.03710316640393,
      "language_loss": 0.78448367,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.8077513,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 3.4798779487609863
    },
    {
      "auxiliary_loss_clip": 0.01186932,
      "auxiliary_loss_mlp": 0.01031842,
      "balance_loss_clip": 0.98036951,
      "balance_loss_mlp": 1.02308047,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.3304332602302154,
      "language_loss": 0.89829361,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.92048138,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 3.5683071613311768
    },
    {
      "auxiliary_loss_clip": 0.01200396,
      "auxiliary_loss_mlp": 0.01034945,
      "balance_loss_clip": 1.02032399,
      "balance_loss_mlp": 1.0255928,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 1.4535808625957451,
      "language_loss": 0.84240103,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.8647545,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.757564067840576
    },
    {
      "auxiliary_loss_clip": 0.01198034,
      "auxiliary_loss_mlp": 0.01037334,
      "balance_loss_clip": 1.02005303,
      "balance_loss_mlp": 1.0283637,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 1.8308670940902592,
      "language_loss": 0.81663263,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.83898628,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.6554839611053467
    },
    {
      "auxiliary_loss_clip": 0.0118741,
      "auxiliary_loss_mlp": 0.01041117,
      "balance_loss_clip": 1.01646101,
      "balance_loss_mlp": 1.03219426,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.5152123317704538,
      "language_loss": 0.77553988,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.7978251,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.7347195148468018
    },
    {
      "auxiliary_loss_clip": 0.01201484,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.02000761,
      "balance_loss_mlp": 1.02054501,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 2.124439441851099,
      "language_loss": 0.82677257,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.84907568,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.7479422092437744
    },
    {
      "auxiliary_loss_clip": 0.01195289,
      "auxiliary_loss_mlp": 0.01031588,
      "balance_loss_clip": 0.94265777,
      "balance_loss_mlp": 1.02306485,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.5451413605392692,
      "language_loss": 0.78337348,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.80564225,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.876481771469116
    },
    {
      "auxiliary_loss_clip": 0.01191169,
      "auxiliary_loss_mlp": 0.01030631,
      "balance_loss_clip": 0.98414069,
      "balance_loss_mlp": 1.02117813,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 1.8627035624138757,
      "language_loss": 0.76304102,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.78525895,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.68685245513916
    },
    {
      "auxiliary_loss_clip": 0.01203123,
      "auxiliary_loss_mlp": 0.01036573,
      "balance_loss_clip": 1.05942023,
      "balance_loss_mlp": 1.02715528,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 1.5550993651939122,
      "language_loss": 0.90174645,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92414343,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 2.68536114692688
    },
    {
      "auxiliary_loss_clip": 0.01202023,
      "auxiliary_loss_mlp": 0.01032657,
      "balance_loss_clip": 1.02288461,
      "balance_loss_mlp": 1.02272666,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 1.99663144475044,
      "language_loss": 0.85118884,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87353563,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.6148226261138916
    },
    {
      "auxiliary_loss_clip": 0.01197212,
      "auxiliary_loss_mlp": 0.0103636,
      "balance_loss_clip": 0.98017716,
      "balance_loss_mlp": 1.02730012,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 2.309097260546068,
      "language_loss": 0.85317504,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.87551069,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.695842742919922
    },
    {
      "auxiliary_loss_clip": 0.01195684,
      "auxiliary_loss_mlp": 0.01031476,
      "balance_loss_clip": 1.01969743,
      "balance_loss_mlp": 1.02264237,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 1.7310032771460764,
      "language_loss": 0.76066709,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.78293872,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 2.6494381427764893
    },
    {
      "auxiliary_loss_clip": 0.01203303,
      "auxiliary_loss_mlp": 0.01032746,
      "balance_loss_clip": 1.05803037,
      "balance_loss_mlp": 1.02337611,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 1.8999650495591898,
      "language_loss": 0.78798342,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.81034386,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 2.6672892570495605
    },
    {
      "auxiliary_loss_clip": 0.01197064,
      "auxiliary_loss_mlp": 0.01125469,
      "balance_loss_clip": 1.0555594,
      "balance_loss_mlp": 0.0,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 1.9952966078674284,
      "language_loss": 0.84689802,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.87012339,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.6238064765930176
    },
    {
      "auxiliary_loss_clip": 0.01199585,
      "auxiliary_loss_mlp": 0.0103474,
      "balance_loss_clip": 1.05743992,
      "balance_loss_mlp": 1.025859,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.7855309865420268,
      "language_loss": 0.79114288,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81348616,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 2.627662420272827
    },
    {
      "auxiliary_loss_clip": 0.0120393,
      "auxiliary_loss_mlp": 0.0112597,
      "balance_loss_clip": 1.01886916,
      "balance_loss_mlp": 0.0,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.1194764008645377,
      "language_loss": 0.80022633,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.82352531,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 2.6651854515075684
    },
    {
      "auxiliary_loss_clip": 0.01087948,
      "auxiliary_loss_mlp": 0.01120423,
      "balance_loss_clip": 1.01842225,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8238776975640061,
      "language_loss": 0.59395629,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61603999,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.1187071800231934
    },
    {
      "auxiliary_loss_clip": 0.01088568,
      "auxiliary_loss_mlp": 0.01002234,
      "balance_loss_clip": 1.01906419,
      "balance_loss_mlp": 0.99955177,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8058883105348184,
      "language_loss": 0.55854583,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57945383,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.222491502761841
    },
    {
      "auxiliary_loss_clip": 0.01198962,
      "auxiliary_loss_mlp": 0.01032444,
      "balance_loss_clip": 1.01793027,
      "balance_loss_mlp": 1.02386737,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 6.774526831813159,
      "language_loss": 0.82485461,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.84716868,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.6238362789154053
    },
    {
      "auxiliary_loss_clip": 0.01194772,
      "auxiliary_loss_mlp": 0.0103554,
      "balance_loss_clip": 1.01851535,
      "balance_loss_mlp": 1.02669466,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 2.4640252447565003,
      "language_loss": 0.87002242,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.89232552,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.7546768188476562
    },
    {
      "auxiliary_loss_clip": 0.01199838,
      "auxiliary_loss_mlp": 0.01024468,
      "balance_loss_clip": 0.97765559,
      "balance_loss_mlp": 1.01594472,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.5027728287343338,
      "language_loss": 0.7567789,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.77902198,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 3.821995973587036
    },
    {
      "auxiliary_loss_clip": 0.011844,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 0.93820846,
      "balance_loss_mlp": 1.02097011,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 2.085303928155695,
      "language_loss": 0.75947392,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.78161538,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 2.8025834560394287
    },
    {
      "auxiliary_loss_clip": 0.01182172,
      "auxiliary_loss_mlp": 0.01035965,
      "balance_loss_clip": 0.89801663,
      "balance_loss_mlp": 1.02677441,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 1.7438086038554792,
      "language_loss": 0.86801016,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.8901915,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 3.8267288208007812
    },
    {
      "auxiliary_loss_clip": 0.01196909,
      "auxiliary_loss_mlp": 0.01025706,
      "balance_loss_clip": 0.98223495,
      "balance_loss_mlp": 1.01626444,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 3.0535873455094658,
      "language_loss": 0.88661242,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.90883851,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.719700574874878
    },
    {
      "auxiliary_loss_clip": 0.01200255,
      "auxiliary_loss_mlp": 0.01031859,
      "balance_loss_clip": 1.05884004,
      "balance_loss_mlp": 1.02332926,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 1.5601685720434828,
      "language_loss": 0.74347872,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76579988,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 2.6738650798797607
    },
    {
      "auxiliary_loss_clip": 0.0118335,
      "auxiliary_loss_mlp": 0.01033856,
      "balance_loss_clip": 0.97913641,
      "balance_loss_mlp": 1.02535069,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.1489554271404656,
      "language_loss": 0.74607283,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.76824492,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.6811118125915527
    },
    {
      "auxiliary_loss_clip": 0.01194281,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 1.01853991,
      "balance_loss_mlp": 1.02049446,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.7961441837694356,
      "language_loss": 0.87870371,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.90093505,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 3.6263251304626465
    },
    {
      "auxiliary_loss_clip": 0.01199843,
      "auxiliary_loss_mlp": 0.01030345,
      "balance_loss_clip": 0.98051113,
      "balance_loss_mlp": 1.02172613,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 1.5961154211126363,
      "language_loss": 0.69360596,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.71590781,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.737704038619995
    },
    {
      "auxiliary_loss_clip": 0.01198841,
      "auxiliary_loss_mlp": 0.01032997,
      "balance_loss_clip": 0.9806757,
      "balance_loss_mlp": 1.02424109,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 1.9920860536207576,
      "language_loss": 0.82864308,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.85096145,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.6216580867767334
    },
    {
      "auxiliary_loss_clip": 0.011877,
      "auxiliary_loss_mlp": 0.01033426,
      "balance_loss_clip": 0.86230457,
      "balance_loss_mlp": 1.02458072,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 1.7054214867910136,
      "language_loss": 0.75522828,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.77743959,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.831193447113037
    },
    {
      "auxiliary_loss_clip": 0.01203877,
      "auxiliary_loss_mlp": 0.01041217,
      "balance_loss_clip": 1.058568,
      "balance_loss_mlp": 1.03237736,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 2.6741834542924985,
      "language_loss": 0.8357023,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.85815328,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.6435112953186035
    },
    {
      "auxiliary_loss_clip": 0.01202779,
      "auxiliary_loss_mlp": 0.01033274,
      "balance_loss_clip": 1.01740968,
      "balance_loss_mlp": 1.0241251,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 2.258516099628326,
      "language_loss": 0.7897504,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.8121109,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 2.6459550857543945
    },
    {
      "auxiliary_loss_clip": 0.01197522,
      "auxiliary_loss_mlp": 0.01034534,
      "balance_loss_clip": 1.01640892,
      "balance_loss_mlp": 1.02506256,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 1.803180139150711,
      "language_loss": 0.81424177,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83656234,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.7046353816986084
    },
    {
      "auxiliary_loss_clip": 0.01203886,
      "auxiliary_loss_mlp": 0.01030633,
      "balance_loss_clip": 1.01892257,
      "balance_loss_mlp": 1.021806,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 1.9154680783688196,
      "language_loss": 0.75901961,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.7813648,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 2.6707499027252197
    },
    {
      "auxiliary_loss_clip": 0.01200647,
      "auxiliary_loss_mlp": 0.0112604,
      "balance_loss_clip": 0.94354659,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 2.9851234007070673,
      "language_loss": 0.73542154,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.75868845,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.7341458797454834
    },
    {
      "auxiliary_loss_clip": 0.011978,
      "auxiliary_loss_mlp": 0.01032905,
      "balance_loss_clip": 0.98421973,
      "balance_loss_mlp": 1.02418458,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 1.83376112509116,
      "language_loss": 0.7352066,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.75751364,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.9351377487182617
    },
    {
      "auxiliary_loss_clip": 0.01182253,
      "auxiliary_loss_mlp": 0.01028726,
      "balance_loss_clip": 0.97873253,
      "balance_loss_mlp": 1.02024412,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 2.1370661553326564,
      "language_loss": 0.75424653,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77635634,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.7488327026367188
    },
    {
      "auxiliary_loss_clip": 0.01200903,
      "auxiliary_loss_mlp": 0.01041188,
      "balance_loss_clip": 0.94368315,
      "balance_loss_mlp": 1.03268814,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 2.7376007849210966,
      "language_loss": 0.66843712,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.69085801,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.8428244590759277
    },
    {
      "auxiliary_loss_clip": 0.01188195,
      "auxiliary_loss_mlp": 0.01034039,
      "balance_loss_clip": 0.97796714,
      "balance_loss_mlp": 1.02503252,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 1.9024505824369993,
      "language_loss": 0.82565558,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.84787792,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.67341685295105
    },
    {
      "auxiliary_loss_clip": 0.01089797,
      "auxiliary_loss_mlp": 0.01003818,
      "balance_loss_clip": 0.98371744,
      "balance_loss_mlp": 1.00113618,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7224573323840551,
      "language_loss": 0.52667248,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54760867,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.3452513217926025
    },
    {
      "auxiliary_loss_clip": 0.01192323,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 0.90482211,
      "balance_loss_mlp": 1.02414596,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.189950922837858,
      "language_loss": 0.95580459,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.97805434,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 2.7326838970184326
    },
    {
      "auxiliary_loss_clip": 0.01195395,
      "auxiliary_loss_mlp": 0.01046136,
      "balance_loss_clip": 1.01813686,
      "balance_loss_mlp": 1.03721368,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 1.9451955803524694,
      "language_loss": 0.83506638,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.85748166,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 2.8346362113952637
    },
    {
      "auxiliary_loss_clip": 0.01199226,
      "auxiliary_loss_mlp": 0.01038015,
      "balance_loss_clip": 1.0588541,
      "balance_loss_mlp": 1.02949178,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 1.741869792010861,
      "language_loss": 0.76684201,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.78921437,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.6264421939849854
    },
    {
      "auxiliary_loss_clip": 0.01094974,
      "auxiliary_loss_mlp": 0.01001526,
      "balance_loss_clip": 0.90775639,
      "balance_loss_mlp": 0.99892753,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.891958062546224,
      "language_loss": 0.60496575,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62593073,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.271017074584961
    },
    {
      "auxiliary_loss_clip": 0.01199217,
      "auxiliary_loss_mlp": 0.01032214,
      "balance_loss_clip": 0.97976428,
      "balance_loss_mlp": 1.0226717,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 1.861240459605536,
      "language_loss": 0.78652728,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.80884165,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.7060179710388184
    },
    {
      "auxiliary_loss_clip": 0.01189742,
      "auxiliary_loss_mlp": 0.01033888,
      "balance_loss_clip": 0.90245372,
      "balance_loss_mlp": 1.02531111,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 2.387189037098938,
      "language_loss": 0.67854887,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.70078516,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.807129383087158
    },
    {
      "auxiliary_loss_clip": 0.01183699,
      "auxiliary_loss_mlp": 0.0103358,
      "balance_loss_clip": 0.94019449,
      "balance_loss_mlp": 1.02534902,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 1.686553075637883,
      "language_loss": 0.67899787,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.70117068,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 3.737443447113037
    },
    {
      "auxiliary_loss_clip": 0.01184917,
      "auxiliary_loss_mlp": 0.01033913,
      "balance_loss_clip": 0.97642338,
      "balance_loss_mlp": 1.0249598,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 1.799030454886866,
      "language_loss": 0.83851135,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.86069965,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 2.635493516921997
    },
    {
      "auxiliary_loss_clip": 0.011839,
      "auxiliary_loss_mlp": 0.01037214,
      "balance_loss_clip": 0.97909337,
      "balance_loss_mlp": 1.02867293,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 1.8665338039362411,
      "language_loss": 0.7673682,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.78957939,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 3.5476083755493164
    },
    {
      "auxiliary_loss_clip": 0.01084369,
      "auxiliary_loss_mlp": 0.01119858,
      "balance_loss_clip": 0.98066092,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8095867177678633,
      "language_loss": 0.60392118,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.62596345,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.3244404792785645
    },
    {
      "auxiliary_loss_clip": 0.01181728,
      "auxiliary_loss_mlp": 0.01026074,
      "balance_loss_clip": 0.93915069,
      "balance_loss_mlp": 1.01790833,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 1.548201592164931,
      "language_loss": 0.7836858,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.80576384,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 2.733959197998047
    },
    {
      "auxiliary_loss_clip": 0.01194484,
      "auxiliary_loss_mlp": 0.01036794,
      "balance_loss_clip": 0.94250405,
      "balance_loss_mlp": 1.02775145,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 1.6895790224775529,
      "language_loss": 0.77563244,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.7979452,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.80682110786438
    },
    {
      "auxiliary_loss_clip": 0.01182208,
      "auxiliary_loss_mlp": 0.01034192,
      "balance_loss_clip": 0.94009244,
      "balance_loss_mlp": 1.02503049,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 2.2897445874552482,
      "language_loss": 0.92209852,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.9442625,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 3.6828696727752686
    },
    {
      "auxiliary_loss_clip": 0.01086573,
      "auxiliary_loss_mlp": 0.01002875,
      "balance_loss_clip": 1.01907933,
      "balance_loss_mlp": 1.00031173,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.8843023758982012,
      "language_loss": 0.63540673,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.6563012,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 4.117944002151489
    },
    {
      "auxiliary_loss_clip": 0.01203542,
      "auxiliary_loss_mlp": 0.01031749,
      "balance_loss_clip": 1.06093907,
      "balance_loss_mlp": 1.02285624,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 1.7908098076440382,
      "language_loss": 0.80954689,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.83189982,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.7048697471618652
    },
    {
      "auxiliary_loss_clip": 0.0119593,
      "auxiliary_loss_mlp": 0.0103319,
      "balance_loss_clip": 1.05814433,
      "balance_loss_mlp": 1.02468479,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 1.9643649504550773,
      "language_loss": 0.86760807,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.88989931,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.5951082706451416
    },
    {
      "auxiliary_loss_clip": 0.01195987,
      "auxiliary_loss_mlp": 0.01025061,
      "balance_loss_clip": 0.86436981,
      "balance_loss_mlp": 1.01650798,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.8596151288676133,
      "language_loss": 0.76918477,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79139531,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.8716142177581787
    },
    {
      "auxiliary_loss_clip": 0.01197471,
      "auxiliary_loss_mlp": 0.01042341,
      "balance_loss_clip": 1.05856359,
      "balance_loss_mlp": 1.03406143,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 1.5932012514595943,
      "language_loss": 0.77533913,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.79773724,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.7141387462615967
    },
    {
      "auxiliary_loss_clip": 0.01183605,
      "auxiliary_loss_mlp": 0.01125702,
      "balance_loss_clip": 0.97728479,
      "balance_loss_mlp": 0.0,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.6859781956319075,
      "language_loss": 0.83870983,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.86180294,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.7679264545440674
    },
    {
      "auxiliary_loss_clip": 0.01204665,
      "auxiliary_loss_mlp": 0.0103686,
      "balance_loss_clip": 0.98017162,
      "balance_loss_mlp": 1.02802026,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 1.7542504107278984,
      "language_loss": 0.66471159,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.68712682,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.711876630783081
    },
    {
      "auxiliary_loss_clip": 0.01195464,
      "auxiliary_loss_mlp": 0.0102986,
      "balance_loss_clip": 1.01944232,
      "balance_loss_mlp": 1.02079976,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 1.490388166614282,
      "language_loss": 0.89256352,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.91481674,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.6855461597442627
    },
    {
      "auxiliary_loss_clip": 0.01200188,
      "auxiliary_loss_mlp": 0.01031874,
      "balance_loss_clip": 1.01879334,
      "balance_loss_mlp": 1.02296352,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 1.9615810896359533,
      "language_loss": 0.73253369,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.75485432,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 2.6582908630371094
    },
    {
      "auxiliary_loss_clip": 0.01194359,
      "auxiliary_loss_mlp": 0.01030018,
      "balance_loss_clip": 0.97717071,
      "balance_loss_mlp": 1.02111948,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 1.6522134433443783,
      "language_loss": 0.84602296,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.8682667,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.726346015930176
    },
    {
      "auxiliary_loss_clip": 0.01200514,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 1.0202204,
      "balance_loss_mlp": 1.02130437,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 2.1450050391690483,
      "language_loss": 0.84981871,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.87212753,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.703623056411743
    },
    {
      "auxiliary_loss_clip": 0.0119907,
      "auxiliary_loss_mlp": 0.01125248,
      "balance_loss_clip": 1.02006364,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.686295672394898,
      "language_loss": 0.85840845,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.88165164,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.717508316040039
    },
    {
      "auxiliary_loss_clip": 0.01197215,
      "auxiliary_loss_mlp": 0.01029499,
      "balance_loss_clip": 1.05764914,
      "balance_loss_mlp": 1.02085578,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.683068503120339,
      "language_loss": 0.84577107,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86803818,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 2.663421154022217
    },
    {
      "auxiliary_loss_clip": 0.01198528,
      "auxiliary_loss_mlp": 0.01029334,
      "balance_loss_clip": 1.06016517,
      "balance_loss_mlp": 1.02110887,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 1.9635200173568403,
      "language_loss": 0.71729219,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.7395708,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 2.6736199855804443
    },
    {
      "auxiliary_loss_clip": 0.01192207,
      "auxiliary_loss_mlp": 0.01027002,
      "balance_loss_clip": 0.94155419,
      "balance_loss_mlp": 1.01848435,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 1.7833376402516696,
      "language_loss": 0.82270896,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84490103,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 2.7116336822509766
    },
    {
      "auxiliary_loss_clip": 0.01195154,
      "auxiliary_loss_mlp": 0.0103338,
      "balance_loss_clip": 0.98239887,
      "balance_loss_mlp": 1.0247438,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 1.5954954704090762,
      "language_loss": 0.77362001,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.79590535,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 2.714571475982666
    },
    {
      "auxiliary_loss_clip": 0.01191054,
      "auxiliary_loss_mlp": 0.01036177,
      "balance_loss_clip": 1.01612616,
      "balance_loss_mlp": 1.02761197,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 2.426899939526433,
      "language_loss": 0.83083409,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.85310644,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.7818562984466553
    },
    {
      "auxiliary_loss_clip": 0.01085529,
      "auxiliary_loss_mlp": 0.01008157,
      "balance_loss_clip": 1.01786137,
      "balance_loss_mlp": 1.00559366,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.7833184841010761,
      "language_loss": 0.55569315,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57663,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.2704920768737793
    },
    {
      "auxiliary_loss_clip": 0.012051,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.02115989,
      "balance_loss_mlp": 1.02041197,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 2.013331526730329,
      "language_loss": 0.82167852,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84401911,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.664198160171509
    },
    {
      "auxiliary_loss_clip": 0.01187811,
      "auxiliary_loss_mlp": 0.01024483,
      "balance_loss_clip": 0.97854173,
      "balance_loss_mlp": 1.01604903,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 1.7642122686250585,
      "language_loss": 0.84019005,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86231291,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 3.778594493865967
    },
    {
      "auxiliary_loss_clip": 0.01190414,
      "auxiliary_loss_mlp": 0.01031604,
      "balance_loss_clip": 0.9831562,
      "balance_loss_mlp": 1.02260983,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 3.943051366435319,
      "language_loss": 0.71078408,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73300421,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.76035737991333
    },
    {
      "auxiliary_loss_clip": 0.0119433,
      "auxiliary_loss_mlp": 0.01032619,
      "balance_loss_clip": 1.02058744,
      "balance_loss_mlp": 1.02397656,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 2.2799740165708924,
      "language_loss": 0.76819414,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.79046363,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 2.6408538818359375
    },
    {
      "auxiliary_loss_clip": 0.011958,
      "auxiliary_loss_mlp": 0.01029082,
      "balance_loss_clip": 1.02010202,
      "balance_loss_mlp": 1.0206418,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 3.4178885190595163,
      "language_loss": 0.72366989,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.74591869,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 3.5793097019195557
    },
    {
      "auxiliary_loss_clip": 0.01196801,
      "auxiliary_loss_mlp": 0.01029633,
      "balance_loss_clip": 0.98136961,
      "balance_loss_mlp": 1.02070987,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 2.0400880962700136,
      "language_loss": 0.88096577,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90323013,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 2.6811673641204834
    },
    {
      "auxiliary_loss_clip": 0.01201001,
      "auxiliary_loss_mlp": 0.01038629,
      "balance_loss_clip": 0.94131458,
      "balance_loss_mlp": 1.02942038,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 1.7311825287224052,
      "language_loss": 0.77624941,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79864573,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.757467746734619
    },
    {
      "auxiliary_loss_clip": 0.01180931,
      "auxiliary_loss_mlp": 0.01037662,
      "balance_loss_clip": 0.93826538,
      "balance_loss_mlp": 1.02866757,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 2.0994033622986383,
      "language_loss": 0.64319289,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.66537881,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 4.603882789611816
    },
    {
      "auxiliary_loss_clip": 0.01085359,
      "auxiliary_loss_mlp": 0.01003039,
      "balance_loss_clip": 1.0183574,
      "balance_loss_mlp": 1.00055933,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7516635338997695,
      "language_loss": 0.56971872,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.5906027,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.180396556854248
    },
    {
      "auxiliary_loss_clip": 0.01203426,
      "auxiliary_loss_mlp": 0.01030815,
      "balance_loss_clip": 1.06111193,
      "balance_loss_mlp": 1.02177262,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 2.153067798562823,
      "language_loss": 0.72043478,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.74277717,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.7090694904327393
    },
    {
      "auxiliary_loss_clip": 0.01194037,
      "auxiliary_loss_mlp": 0.01029987,
      "balance_loss_clip": 1.01770914,
      "balance_loss_mlp": 1.02160597,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 1.7543987475681038,
      "language_loss": 0.69625747,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.71849769,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.6820385456085205
    },
    {
      "auxiliary_loss_clip": 0.01202776,
      "auxiliary_loss_mlp": 0.01031171,
      "balance_loss_clip": 0.98441553,
      "balance_loss_mlp": 1.02291,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 1.7048653599461716,
      "language_loss": 0.88473892,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.90707839,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.740013599395752
    },
    {
      "auxiliary_loss_clip": 0.01194058,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 0.98004907,
      "balance_loss_mlp": 1.02044749,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 1.7195432639260475,
      "language_loss": 0.79806775,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.82029665,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.7858870029449463
    },
    {
      "auxiliary_loss_clip": 0.01203258,
      "auxiliary_loss_mlp": 0.01032711,
      "balance_loss_clip": 1.02197278,
      "balance_loss_mlp": 1.02369845,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.154811092128439,
      "language_loss": 0.71981919,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.74217892,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 2.6849985122680664
    },
    {
      "auxiliary_loss_clip": 0.01185756,
      "auxiliary_loss_mlp": 0.01033808,
      "balance_loss_clip": 0.93976641,
      "balance_loss_mlp": 1.02535021,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 2.8766111222829664,
      "language_loss": 0.8975966,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.91979218,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.7670774459838867
    },
    {
      "auxiliary_loss_clip": 0.01090693,
      "auxiliary_loss_mlp": 0.01007985,
      "balance_loss_clip": 0.9475863,
      "balance_loss_mlp": 1.00512362,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.7977846517182942,
      "language_loss": 0.62833011,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64931691,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.3349130153656006
    },
    {
      "auxiliary_loss_clip": 0.01195776,
      "auxiliary_loss_mlp": 0.01032913,
      "balance_loss_clip": 0.94180918,
      "balance_loss_mlp": 1.02440178,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.2351878845169875,
      "language_loss": 0.81562185,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.83790869,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 2.7295897006988525
    },
    {
      "auxiliary_loss_clip": 0.01201361,
      "auxiliary_loss_mlp": 0.01042277,
      "balance_loss_clip": 1.06055629,
      "balance_loss_mlp": 1.03391421,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 1.7122434061298626,
      "language_loss": 0.75414789,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77658427,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 2.593919515609741
    },
    {
      "auxiliary_loss_clip": 0.01194825,
      "auxiliary_loss_mlp": 0.01025402,
      "balance_loss_clip": 1.02019,
      "balance_loss_mlp": 1.01739669,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.4225600617529346,
      "language_loss": 0.81546628,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83766854,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.7131662368774414
    },
    {
      "auxiliary_loss_clip": 0.01177466,
      "auxiliary_loss_mlp": 0.01038775,
      "balance_loss_clip": 0.93740714,
      "balance_loss_mlp": 1.02948308,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 3.2845327131746864,
      "language_loss": 0.74541634,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.76757872,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.706059455871582
    },
    {
      "auxiliary_loss_clip": 0.01189509,
      "auxiliary_loss_mlp": 0.01031076,
      "balance_loss_clip": 0.90619767,
      "balance_loss_mlp": 1.02283323,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 1.9173520242765674,
      "language_loss": 0.79864359,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.82084948,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 2.7146835327148438
    },
    {
      "auxiliary_loss_clip": 0.01191554,
      "auxiliary_loss_mlp": 0.01033965,
      "balance_loss_clip": 0.94063061,
      "balance_loss_mlp": 1.02455318,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 1.9788768879284193,
      "language_loss": 0.77950794,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80176318,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 2.7102324962615967
    },
    {
      "auxiliary_loss_clip": 0.01185741,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 0.97835815,
      "balance_loss_mlp": 1.01902246,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 1.9890600127717268,
      "language_loss": 0.73007387,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.75220799,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 2.68745756149292
    },
    {
      "auxiliary_loss_clip": 0.01194624,
      "auxiliary_loss_mlp": 0.01023792,
      "balance_loss_clip": 1.01984215,
      "balance_loss_mlp": 1.01538765,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 1.664397843604086,
      "language_loss": 0.75198913,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.77417332,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 2.838449478149414
    },
    {
      "auxiliary_loss_clip": 0.01193589,
      "auxiliary_loss_mlp": 0.01040763,
      "balance_loss_clip": 0.94269985,
      "balance_loss_mlp": 1.03244185,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 1.9097852825658226,
      "language_loss": 0.80670828,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82905179,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.7118124961853027
    },
    {
      "auxiliary_loss_clip": 0.01200867,
      "auxiliary_loss_mlp": 0.01031797,
      "balance_loss_clip": 1.02104318,
      "balance_loss_mlp": 1.02227807,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 3.760761831012303,
      "language_loss": 0.70347941,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.72580606,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.6654884815216064
    },
    {
      "auxiliary_loss_clip": 0.01195786,
      "auxiliary_loss_mlp": 0.01031125,
      "balance_loss_clip": 1.02146602,
      "balance_loss_mlp": 1.02267933,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 1.6671979682321747,
      "language_loss": 0.75668919,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.77895832,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.6455044746398926
    },
    {
      "auxiliary_loss_clip": 0.011972,
      "auxiliary_loss_mlp": 0.01124841,
      "balance_loss_clip": 0.97926366,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 1.9846249834741,
      "language_loss": 0.84146613,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.86468649,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 3.7111761569976807
    },
    {
      "auxiliary_loss_clip": 0.01192127,
      "auxiliary_loss_mlp": 0.01030708,
      "balance_loss_clip": 1.01549208,
      "balance_loss_mlp": 1.02270365,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 2.5226064444318586,
      "language_loss": 0.91256666,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.93479502,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.6396546363830566
    },
    {
      "auxiliary_loss_clip": 0.01186882,
      "auxiliary_loss_mlp": 0.01031242,
      "balance_loss_clip": 0.86393988,
      "balance_loss_mlp": 1.02305269,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 2.235544131876833,
      "language_loss": 0.82686931,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.84905052,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 2.782479763031006
    },
    {
      "auxiliary_loss_clip": 0.01200883,
      "auxiliary_loss_mlp": 0.01037241,
      "balance_loss_clip": 0.98083717,
      "balance_loss_mlp": 1.02917087,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.6337024468922967,
      "language_loss": 0.70683193,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.72921318,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 3.798234462738037
    },
    {
      "auxiliary_loss_clip": 0.01194434,
      "auxiliary_loss_mlp": 0.01034568,
      "balance_loss_clip": 1.05738473,
      "balance_loss_mlp": 1.02660525,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 1.7373863016908948,
      "language_loss": 0.80352914,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82581925,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 2.5691184997558594
    },
    {
      "auxiliary_loss_clip": 0.01196997,
      "auxiliary_loss_mlp": 0.01035981,
      "balance_loss_clip": 1.05933797,
      "balance_loss_mlp": 1.02797627,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 1.8396258265062924,
      "language_loss": 0.75437248,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77670228,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 2.634591579437256
    },
    {
      "auxiliary_loss_clip": 0.01197101,
      "auxiliary_loss_mlp": 0.01030832,
      "balance_loss_clip": 1.0190649,
      "balance_loss_mlp": 1.02238619,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 1.7809910870335435,
      "language_loss": 0.7972883,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81956756,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 4.747631311416626
    },
    {
      "auxiliary_loss_clip": 0.01086249,
      "auxiliary_loss_mlp": 0.01011942,
      "balance_loss_clip": 1.01878035,
      "balance_loss_mlp": 1.00930774,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.7748990791787259,
      "language_loss": 0.55257213,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57355404,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 3.3284995555877686
    },
    {
      "auxiliary_loss_clip": 0.01189919,
      "auxiliary_loss_mlp": 0.01025028,
      "balance_loss_clip": 0.97952664,
      "balance_loss_mlp": 1.01686192,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.5850539553902196,
      "language_loss": 0.79900289,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82115239,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.765580654144287
    },
    {
      "auxiliary_loss_clip": 0.01196549,
      "auxiliary_loss_mlp": 0.0103023,
      "balance_loss_clip": 1.05725384,
      "balance_loss_mlp": 1.02203393,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 1.5364599936901047,
      "language_loss": 0.84188241,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86415017,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.710705280303955
    },
    {
      "auxiliary_loss_clip": 0.01196904,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.05700362,
      "balance_loss_mlp": 1.02323866,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 1.870088524686712,
      "language_loss": 0.84383696,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86612546,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.67006516456604
    },
    {
      "auxiliary_loss_clip": 0.01179412,
      "auxiliary_loss_mlp": 0.01029776,
      "balance_loss_clip": 0.90056634,
      "balance_loss_mlp": 1.02109718,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.739775049660078,
      "language_loss": 0.86481071,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.88690263,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.8048036098480225
    },
    {
      "auxiliary_loss_clip": 0.01193325,
      "auxiliary_loss_mlp": 0.01027454,
      "balance_loss_clip": 0.94338584,
      "balance_loss_mlp": 1.01928854,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 1.96339292410903,
      "language_loss": 0.72942346,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75163126,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 2.702547550201416
    },
    {
      "auxiliary_loss_clip": 0.01199214,
      "auxiliary_loss_mlp": 0.01030583,
      "balance_loss_clip": 1.05777681,
      "balance_loss_mlp": 1.0220958,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 1.8516342711629137,
      "language_loss": 0.84575307,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.86805105,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.6240601539611816
    },
    {
      "auxiliary_loss_clip": 0.01191483,
      "auxiliary_loss_mlp": 0.01035816,
      "balance_loss_clip": 1.01771986,
      "balance_loss_mlp": 1.02668452,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 1.8910099001854472,
      "language_loss": 0.79742664,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.81969965,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.645326852798462
    },
    {
      "auxiliary_loss_clip": 0.01199631,
      "auxiliary_loss_mlp": 0.0104089,
      "balance_loss_clip": 0.98187178,
      "balance_loss_mlp": 1.03233659,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 5.413212469594398,
      "language_loss": 0.65971625,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.6821214,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 2.820629835128784
    },
    {
      "auxiliary_loss_clip": 0.01198407,
      "auxiliary_loss_mlp": 0.01030434,
      "balance_loss_clip": 0.9778353,
      "balance_loss_mlp": 1.0213387,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 1.9563002142953658,
      "language_loss": 0.84830201,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87059033,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 2.7471227645874023
    },
    {
      "auxiliary_loss_clip": 0.01190666,
      "auxiliary_loss_mlp": 0.0102678,
      "balance_loss_clip": 1.01504803,
      "balance_loss_mlp": 1.01783919,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 2.0359078980004637,
      "language_loss": 0.86778444,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.88995886,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 2.721558094024658
    },
    {
      "auxiliary_loss_clip": 0.01196564,
      "auxiliary_loss_mlp": 0.01030865,
      "balance_loss_clip": 0.94033396,
      "balance_loss_mlp": 1.02203751,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 1.5549195859291973,
      "language_loss": 0.75466108,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77693534,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.8533473014831543
    },
    {
      "auxiliary_loss_clip": 0.01197401,
      "auxiliary_loss_mlp": 0.01036742,
      "balance_loss_clip": 1.01911426,
      "balance_loss_mlp": 1.0276283,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.7845738256338013,
      "language_loss": 0.69382358,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.71616507,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.737701654434204
    },
    {
      "auxiliary_loss_clip": 0.01190344,
      "auxiliary_loss_mlp": 0.0102788,
      "balance_loss_clip": 1.01624072,
      "balance_loss_mlp": 1.01978528,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 1.7064826070102344,
      "language_loss": 0.80385232,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.82603455,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 2.7543203830718994
    },
    {
      "auxiliary_loss_clip": 0.01197057,
      "auxiliary_loss_mlp": 0.01037229,
      "balance_loss_clip": 0.98150873,
      "balance_loss_mlp": 1.02868211,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 1.6998746912008595,
      "language_loss": 0.70746797,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.72981083,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 2.679203510284424
    },
    {
      "auxiliary_loss_clip": 0.01196231,
      "auxiliary_loss_mlp": 0.01033085,
      "balance_loss_clip": 0.94219685,
      "balance_loss_mlp": 1.02482414,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 3.5446110951012826,
      "language_loss": 0.73851812,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.76081127,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.7140355110168457
    },
    {
      "auxiliary_loss_clip": 0.01198163,
      "auxiliary_loss_mlp": 0.01030047,
      "balance_loss_clip": 1.05356455,
      "balance_loss_mlp": 1.02063572,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.1024226532080017,
      "language_loss": 0.91018236,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.93246436,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.6393237113952637
    },
    {
      "auxiliary_loss_clip": 0.01200992,
      "auxiliary_loss_mlp": 0.01032295,
      "balance_loss_clip": 1.05821848,
      "balance_loss_mlp": 1.02296066,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.6390615168163505,
      "language_loss": 0.81036752,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.83270037,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.658038377761841
    },
    {
      "auxiliary_loss_clip": 0.01193297,
      "auxiliary_loss_mlp": 0.01042533,
      "balance_loss_clip": 0.9468689,
      "balance_loss_mlp": 1.03415906,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.749257857649044,
      "language_loss": 0.87269795,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89505625,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.686155080795288
    },
    {
      "auxiliary_loss_clip": 0.01195361,
      "auxiliary_loss_mlp": 0.01039692,
      "balance_loss_clip": 0.98029429,
      "balance_loss_mlp": 1.03025031,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.6854328602852193,
      "language_loss": 0.85234243,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87469292,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 3.65584397315979
    },
    {
      "auxiliary_loss_clip": 0.01195271,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.01827228,
      "balance_loss_mlp": 1.02011275,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 2.359658496424881,
      "language_loss": 0.6121242,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.63436484,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.6777963638305664
    },
    {
      "auxiliary_loss_clip": 0.0118524,
      "auxiliary_loss_mlp": 0.01034355,
      "balance_loss_clip": 0.9032644,
      "balance_loss_mlp": 1.02568889,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 2.0610533526692003,
      "language_loss": 0.73066628,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.75286222,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 2.857572555541992
    },
    {
      "auxiliary_loss_clip": 0.01196123,
      "auxiliary_loss_mlp": 0.01029095,
      "balance_loss_clip": 0.97963762,
      "balance_loss_mlp": 1.02078605,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 2.2413979834020874,
      "language_loss": 0.84736955,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.86962175,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 3.7115864753723145
    },
    {
      "auxiliary_loss_clip": 0.01196548,
      "auxiliary_loss_mlp": 0.01026972,
      "balance_loss_clip": 1.05511773,
      "balance_loss_mlp": 1.01833487,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 2.3257022300253967,
      "language_loss": 0.71807748,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.74031264,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 2.556215763092041
    },
    {
      "auxiliary_loss_clip": 0.01198253,
      "auxiliary_loss_mlp": 0.01034874,
      "balance_loss_clip": 1.05557382,
      "balance_loss_mlp": 1.02597463,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 2.572914387148002,
      "language_loss": 0.78492302,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.80725425,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 2.6347668170928955
    },
    {
      "auxiliary_loss_clip": 0.01197741,
      "auxiliary_loss_mlp": 0.01040447,
      "balance_loss_clip": 1.02139688,
      "balance_loss_mlp": 1.03094041,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 1.9334589778083475,
      "language_loss": 0.8345378,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.85691965,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 3.6229937076568604
    },
    {
      "auxiliary_loss_clip": 0.01200645,
      "auxiliary_loss_mlp": 0.01029631,
      "balance_loss_clip": 1.05976355,
      "balance_loss_mlp": 1.02054715,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 2.5218561999818405,
      "language_loss": 0.74158597,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.76388872,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 3.726641893386841
    },
    {
      "auxiliary_loss_clip": 0.01189819,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 0.97678602,
      "balance_loss_mlp": 1.0180769,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 1.8579255477499392,
      "language_loss": 0.85841101,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.8805747,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 2.709372043609619
    },
    {
      "auxiliary_loss_clip": 0.01182582,
      "auxiliary_loss_mlp": 0.01033225,
      "balance_loss_clip": 0.90206641,
      "balance_loss_mlp": 1.02481508,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 1.5859411314781993,
      "language_loss": 0.77633214,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.79849029,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.7575464248657227
    },
    {
      "auxiliary_loss_clip": 0.01200438,
      "auxiliary_loss_mlp": 0.0102992,
      "balance_loss_clip": 1.02245033,
      "balance_loss_mlp": 1.02106881,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 2.0633041971329504,
      "language_loss": 0.84331936,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86562294,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.664201498031616
    },
    {
      "auxiliary_loss_clip": 0.01090326,
      "auxiliary_loss_mlp": 0.01003041,
      "balance_loss_clip": 0.98321128,
      "balance_loss_mlp": 1.00033462,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6383996561825012,
      "language_loss": 0.50333011,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52426374,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.2691726684570312
    },
    {
      "auxiliary_loss_clip": 0.01196345,
      "auxiliary_loss_mlp": 0.01031467,
      "balance_loss_clip": 1.05682659,
      "balance_loss_mlp": 1.02364135,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 2.009581585115112,
      "language_loss": 0.78869289,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.81097096,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.573209047317505
    },
    {
      "auxiliary_loss_clip": 0.0119732,
      "auxiliary_loss_mlp": 0.01032184,
      "balance_loss_clip": 0.9770245,
      "balance_loss_mlp": 1.02344012,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.07994636163549,
      "language_loss": 0.81120968,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83350468,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.7539114952087402
    },
    {
      "auxiliary_loss_clip": 0.01193325,
      "auxiliary_loss_mlp": 0.01033175,
      "balance_loss_clip": 1.01784921,
      "balance_loss_mlp": 1.0248785,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 1.4697515232092626,
      "language_loss": 0.80236,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82462502,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.6878631114959717
    },
    {
      "auxiliary_loss_clip": 0.01198374,
      "auxiliary_loss_mlp": 0.01125163,
      "balance_loss_clip": 1.0580194,
      "balance_loss_mlp": 0.0,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 1.965101176638041,
      "language_loss": 0.86651605,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.88975143,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.6179237365722656
    },
    {
      "auxiliary_loss_clip": 0.01091056,
      "auxiliary_loss_mlp": 0.01000251,
      "balance_loss_clip": 0.98306119,
      "balance_loss_mlp": 0.99760437,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6717828506562884,
      "language_loss": 0.55734861,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.57826167,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.282228708267212
    },
    {
      "auxiliary_loss_clip": 0.0119946,
      "auxiliary_loss_mlp": 0.01032015,
      "balance_loss_clip": 1.01714945,
      "balance_loss_mlp": 1.02357495,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.8608385252823316,
      "language_loss": 0.88688916,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.90920389,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.6531803607940674
    },
    {
      "auxiliary_loss_clip": 0.01180544,
      "auxiliary_loss_mlp": 0.01033126,
      "balance_loss_clip": 0.9357456,
      "balance_loss_mlp": 1.02413213,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 1.7597020215364407,
      "language_loss": 0.77424413,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79638082,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.698190927505493
    },
    {
      "auxiliary_loss_clip": 0.01197339,
      "auxiliary_loss_mlp": 0.01032177,
      "balance_loss_clip": 1.0583632,
      "balance_loss_mlp": 1.02425516,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.4892516376177787,
      "language_loss": 0.71901882,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74131399,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.6661019325256348
    },
    {
      "auxiliary_loss_clip": 0.01196202,
      "auxiliary_loss_mlp": 0.01034614,
      "balance_loss_clip": 1.05750895,
      "balance_loss_mlp": 1.02657354,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 1.5723743600642301,
      "language_loss": 0.87928534,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.90159345,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 2.648273468017578
    },
    {
      "auxiliary_loss_clip": 0.0119587,
      "auxiliary_loss_mlp": 0.01030259,
      "balance_loss_clip": 0.97997952,
      "balance_loss_mlp": 1.02202153,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 1.74167743693137,
      "language_loss": 0.75357115,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77583247,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 2.6991007328033447
    },
    {
      "auxiliary_loss_clip": 0.01182362,
      "auxiliary_loss_mlp": 0.0102634,
      "balance_loss_clip": 0.90262091,
      "balance_loss_mlp": 1.01744068,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 1.7706247625845146,
      "language_loss": 0.85265064,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87473768,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 2.759361743927002
    },
    {
      "auxiliary_loss_clip": 0.01190646,
      "auxiliary_loss_mlp": 0.01124953,
      "balance_loss_clip": 0.97940886,
      "balance_loss_mlp": 0.0,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.7870039982277781,
      "language_loss": 0.79932678,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.82248271,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.664414882659912
    },
    {
      "auxiliary_loss_clip": 0.01087181,
      "auxiliary_loss_mlp": 0.01004117,
      "balance_loss_clip": 0.98063147,
      "balance_loss_mlp": 1.0014466,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.8707159103910906,
      "language_loss": 0.57551908,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59643209,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.3206422328948975
    },
    {
      "auxiliary_loss_clip": 0.01193937,
      "auxiliary_loss_mlp": 0.010296,
      "balance_loss_clip": 0.94420063,
      "balance_loss_mlp": 1.02179778,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.5545335511991756,
      "language_loss": 0.71211141,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73434675,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.817732334136963
    },
    {
      "auxiliary_loss_clip": 0.01194819,
      "auxiliary_loss_mlp": 0.01034205,
      "balance_loss_clip": 1.01642358,
      "balance_loss_mlp": 1.02609837,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 1.745228551633673,
      "language_loss": 0.66884208,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.69113231,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 3.77008056640625
    },
    {
      "auxiliary_loss_clip": 0.01184912,
      "auxiliary_loss_mlp": 0.0103731,
      "balance_loss_clip": 0.93848163,
      "balance_loss_mlp": 1.02845871,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.7965958378046227,
      "language_loss": 0.79929829,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.82152051,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.8337464332580566
    },
    {
      "auxiliary_loss_clip": 0.01195278,
      "auxiliary_loss_mlp": 0.01036267,
      "balance_loss_clip": 0.98377621,
      "balance_loss_mlp": 1.02749336,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.504145140112092,
      "language_loss": 0.75797164,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.78028703,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.857879638671875
    },
    {
      "auxiliary_loss_clip": 0.0120046,
      "auxiliary_loss_mlp": 0.01030675,
      "balance_loss_clip": 1.01840186,
      "balance_loss_mlp": 1.02208602,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.7635046955147031,
      "language_loss": 0.78081381,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.8031252,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.728201389312744
    },
    {
      "auxiliary_loss_clip": 0.01182389,
      "auxiliary_loss_mlp": 0.01033727,
      "balance_loss_clip": 0.97740561,
      "balance_loss_mlp": 1.02475011,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 3.2399400261509443,
      "language_loss": 0.61047643,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.63263762,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 3.6654562950134277
    },
    {
      "auxiliary_loss_clip": 0.01186275,
      "auxiliary_loss_mlp": 0.01031008,
      "balance_loss_clip": 0.93572313,
      "balance_loss_mlp": 1.02287245,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 2.1808756922742,
      "language_loss": 0.81760663,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.8397795,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.7309257984161377
    },
    {
      "auxiliary_loss_clip": 0.0118454,
      "auxiliary_loss_mlp": 0.01027628,
      "balance_loss_clip": 0.97701859,
      "balance_loss_mlp": 1.01963496,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 1.9627602703199116,
      "language_loss": 0.70207012,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.72419184,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 3.6542177200317383
    },
    {
      "auxiliary_loss_clip": 0.01194601,
      "auxiliary_loss_mlp": 0.01034146,
      "balance_loss_clip": 0.98197949,
      "balance_loss_mlp": 1.02636743,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 2.1950260729291085,
      "language_loss": 0.84111035,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86339784,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 3.7023699283599854
    },
    {
      "auxiliary_loss_clip": 0.01192125,
      "auxiliary_loss_mlp": 0.01124475,
      "balance_loss_clip": 0.94313216,
      "balance_loss_mlp": 0.0,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.2933536426177705,
      "language_loss": 0.76075941,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.78392541,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 2.7150204181671143
    },
    {
      "auxiliary_loss_clip": 0.01190882,
      "auxiliary_loss_mlp": 0.01033077,
      "balance_loss_clip": 0.97689497,
      "balance_loss_mlp": 1.02449989,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 3.1662062422911537,
      "language_loss": 0.77823305,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80047262,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.713268280029297
    },
    {
      "auxiliary_loss_clip": 0.01187949,
      "auxiliary_loss_mlp": 0.0102703,
      "balance_loss_clip": 0.97641897,
      "balance_loss_mlp": 1.01892996,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 1.7228917073530667,
      "language_loss": 0.74400944,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.76615924,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.9181666374206543
    },
    {
      "auxiliary_loss_clip": 0.01180899,
      "auxiliary_loss_mlp": 0.01025546,
      "balance_loss_clip": 0.93735993,
      "balance_loss_mlp": 1.01810741,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 1.880682831894626,
      "language_loss": 0.74633372,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.76839817,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.7917380332946777
    },
    {
      "auxiliary_loss_clip": 0.01188949,
      "auxiliary_loss_mlp": 0.01029325,
      "balance_loss_clip": 0.86323774,
      "balance_loss_mlp": 1.02024162,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 2.197110621056796,
      "language_loss": 0.87487769,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.89706051,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.7609450817108154
    },
    {
      "auxiliary_loss_clip": 0.01202429,
      "auxiliary_loss_mlp": 0.01035581,
      "balance_loss_clip": 0.98145592,
      "balance_loss_mlp": 1.0261749,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 1.8660495748991532,
      "language_loss": 0.78854662,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.81092674,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.7068049907684326
    },
    {
      "auxiliary_loss_clip": 0.01192064,
      "auxiliary_loss_mlp": 0.01124783,
      "balance_loss_clip": 1.01606178,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 1.7912303299043557,
      "language_loss": 0.80000329,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.82317179,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.61887788772583
    },
    {
      "auxiliary_loss_clip": 0.01186003,
      "auxiliary_loss_mlp": 0.01034991,
      "balance_loss_clip": 0.90004623,
      "balance_loss_mlp": 1.02574635,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 2.1794337024928425,
      "language_loss": 0.8110069,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.83321679,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.7741012573242188
    },
    {
      "auxiliary_loss_clip": 0.01199214,
      "auxiliary_loss_mlp": 0.01029925,
      "balance_loss_clip": 0.97606856,
      "balance_loss_mlp": 1.02178335,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 1.9296677824740698,
      "language_loss": 0.86394858,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88623995,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 2.712097406387329
    },
    {
      "auxiliary_loss_clip": 0.0118488,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 0.93921101,
      "balance_loss_mlp": 1.02388835,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 3.379061891487731,
      "language_loss": 0.66329408,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.68546265,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.7502083778381348
    },
    {
      "auxiliary_loss_clip": 0.01189477,
      "auxiliary_loss_mlp": 0.0102892,
      "balance_loss_clip": 1.01728737,
      "balance_loss_mlp": 1.02007425,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 1.964659122070503,
      "language_loss": 0.90891373,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93109775,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.7354793548583984
    },
    {
      "auxiliary_loss_clip": 0.01194279,
      "auxiliary_loss_mlp": 0.01125506,
      "balance_loss_clip": 1.01874566,
      "balance_loss_mlp": 0.0,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 1.909747061371216,
      "language_loss": 0.7025193,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.72571719,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.770869731903076
    },
    {
      "auxiliary_loss_clip": 0.01196101,
      "auxiliary_loss_mlp": 0.01024923,
      "balance_loss_clip": 1.05581081,
      "balance_loss_mlp": 1.01655447,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.349619804447457,
      "language_loss": 0.78272808,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80493832,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 2.6663854122161865
    },
    {
      "auxiliary_loss_clip": 0.01193858,
      "auxiliary_loss_mlp": 0.01030431,
      "balance_loss_clip": 1.05316722,
      "balance_loss_mlp": 1.02174091,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.1669050792432505,
      "language_loss": 0.86457348,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.88681638,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 2.656998634338379
    },
    {
      "auxiliary_loss_clip": 0.01182493,
      "auxiliary_loss_mlp": 0.01033974,
      "balance_loss_clip": 0.97803205,
      "balance_loss_mlp": 1.02582002,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 1.5772663886151435,
      "language_loss": 0.8029294,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.8250941,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.710123062133789
    },
    {
      "auxiliary_loss_clip": 0.01193533,
      "auxiliary_loss_mlp": 0.01027512,
      "balance_loss_clip": 1.05342412,
      "balance_loss_mlp": 1.0193882,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 1.919817962933458,
      "language_loss": 0.85802698,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.88023746,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.6138858795166016
    },
    {
      "auxiliary_loss_clip": 0.01190614,
      "auxiliary_loss_mlp": 0.01031063,
      "balance_loss_clip": 1.01716971,
      "balance_loss_mlp": 1.02320743,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 1.9170719495322754,
      "language_loss": 0.71216953,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73438632,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.631021022796631
    },
    {
      "auxiliary_loss_clip": 0.01186068,
      "auxiliary_loss_mlp": 0.01032445,
      "balance_loss_clip": 0.90249717,
      "balance_loss_mlp": 1.02399945,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 1.9533140873575185,
      "language_loss": 0.80961251,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.83179772,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 2.730018138885498
    },
    {
      "auxiliary_loss_clip": 0.01202881,
      "auxiliary_loss_mlp": 0.01125231,
      "balance_loss_clip": 0.98228061,
      "balance_loss_mlp": 0.0,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.0378225880264136,
      "language_loss": 0.75198734,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.77526844,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 3.7310445308685303
    },
    {
      "auxiliary_loss_clip": 0.01198068,
      "auxiliary_loss_mlp": 0.0103368,
      "balance_loss_clip": 1.05630434,
      "balance_loss_mlp": 1.02513909,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 2.0320906919119115,
      "language_loss": 0.82159889,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84391636,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.635007619857788
    },
    {
      "auxiliary_loss_clip": 0.01195205,
      "auxiliary_loss_mlp": 0.01025596,
      "balance_loss_clip": 1.05406094,
      "balance_loss_mlp": 1.01725698,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 1.9953414215602232,
      "language_loss": 0.78373992,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80594796,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 2.655689001083374
    },
    {
      "auxiliary_loss_clip": 0.01193003,
      "auxiliary_loss_mlp": 0.01032112,
      "balance_loss_clip": 0.94114506,
      "balance_loss_mlp": 1.02371943,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 1.6820078048522606,
      "language_loss": 0.78085613,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80310726,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 3.787700653076172
    },
    {
      "auxiliary_loss_clip": 0.0118457,
      "auxiliary_loss_mlp": 0.01037601,
      "balance_loss_clip": 0.86225003,
      "balance_loss_mlp": 1.02874947,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.5899938357909527,
      "language_loss": 0.79829812,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.82051986,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 2.7607080936431885
    },
    {
      "auxiliary_loss_clip": 0.01084676,
      "auxiliary_loss_mlp": 0.01003643,
      "balance_loss_clip": 0.98152179,
      "balance_loss_mlp": 1.00106812,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.759169586110985,
      "language_loss": 0.61579883,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63668203,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.3506836891174316
    },
    {
      "auxiliary_loss_clip": 0.01201802,
      "auxiliary_loss_mlp": 0.01036598,
      "balance_loss_clip": 1.05748868,
      "balance_loss_mlp": 1.02779436,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 2.459289773594848,
      "language_loss": 0.76548809,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.78787208,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.6427016258239746
    },
    {
      "auxiliary_loss_clip": 0.01187291,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 1.01428306,
      "balance_loss_mlp": 1.02195787,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 2.136211264194758,
      "language_loss": 0.83471322,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.85688931,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 4.415686130523682
    },
    {
      "auxiliary_loss_clip": 0.01192205,
      "auxiliary_loss_mlp": 0.01031826,
      "balance_loss_clip": 1.0177542,
      "balance_loss_mlp": 1.02348161,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 1.8049544115852616,
      "language_loss": 0.72334582,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.7455861,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.693248987197876
    },
    {
      "auxiliary_loss_clip": 0.01174429,
      "auxiliary_loss_mlp": 0.0103515,
      "balance_loss_clip": 0.93453228,
      "balance_loss_mlp": 1.0251658,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 1.7236092596995947,
      "language_loss": 0.78509808,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80719382,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.7173678874969482
    },
    {
      "auxiliary_loss_clip": 0.01199264,
      "auxiliary_loss_mlp": 0.01036608,
      "balance_loss_clip": 0.94473702,
      "balance_loss_mlp": 1.02727962,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 2.0272307582098454,
      "language_loss": 0.72586441,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.74822313,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.766071319580078
    },
    {
      "auxiliary_loss_clip": 0.01090825,
      "auxiliary_loss_mlp": 0.01119865,
      "balance_loss_clip": 0.98368758,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.828373911334382,
      "language_loss": 0.62414557,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.64625245,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.3409008979797363
    },
    {
      "auxiliary_loss_clip": 0.01190062,
      "auxiliary_loss_mlp": 0.01031205,
      "balance_loss_clip": 0.90003085,
      "balance_loss_mlp": 1.0222944,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 3.7274351452211554,
      "language_loss": 0.8374548,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.85966742,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.785278558731079
    },
    {
      "auxiliary_loss_clip": 0.01197842,
      "auxiliary_loss_mlp": 0.010388,
      "balance_loss_clip": 1.01897097,
      "balance_loss_mlp": 1.02985954,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 1.9947455448575917,
      "language_loss": 0.94827366,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.97064006,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.658219337463379
    },
    {
      "auxiliary_loss_clip": 0.01196263,
      "auxiliary_loss_mlp": 0.01032846,
      "balance_loss_clip": 0.98311043,
      "balance_loss_mlp": 1.02429271,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 1.3163312730859518,
      "language_loss": 0.84747887,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.86976999,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.6850943565368652
    },
    {
      "auxiliary_loss_clip": 0.01198614,
      "auxiliary_loss_mlp": 0.01033789,
      "balance_loss_clip": 1.0571382,
      "balance_loss_mlp": 1.02546883,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 1.985300768667301,
      "language_loss": 0.83502567,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.85734975,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.6231491565704346
    },
    {
      "auxiliary_loss_clip": 0.01197943,
      "auxiliary_loss_mlp": 0.01030164,
      "balance_loss_clip": 0.98022527,
      "balance_loss_mlp": 1.02143228,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 1.8854633026931265,
      "language_loss": 0.68241334,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.70469439,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.7793338298797607
    },
    {
      "auxiliary_loss_clip": 0.01188976,
      "auxiliary_loss_mlp": 0.01031179,
      "balance_loss_clip": 0.86365747,
      "balance_loss_mlp": 1.02322817,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 2.727639371690942,
      "language_loss": 0.62202942,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.6442309,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.729443311691284
    },
    {
      "auxiliary_loss_clip": 0.01184972,
      "auxiliary_loss_mlp": 0.01033032,
      "balance_loss_clip": 1.01625454,
      "balance_loss_mlp": 1.02388835,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.471665612781208,
      "language_loss": 0.83283341,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85501349,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.763153076171875
    },
    {
      "auxiliary_loss_clip": 0.01087439,
      "auxiliary_loss_mlp": 0.0100611,
      "balance_loss_clip": 0.94511956,
      "balance_loss_mlp": 1.00354719,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7804041468815756,
      "language_loss": 0.59496146,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61589694,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 3.139488458633423
    },
    {
      "auxiliary_loss_clip": 0.01191983,
      "auxiliary_loss_mlp": 0.01035735,
      "balance_loss_clip": 0.97913557,
      "balance_loss_mlp": 1.02725875,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 2.364686317776599,
      "language_loss": 0.67776078,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.70003796,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 2.798790693283081
    },
    {
      "auxiliary_loss_clip": 0.01187548,
      "auxiliary_loss_mlp": 0.01023763,
      "balance_loss_clip": 1.01675463,
      "balance_loss_mlp": 1.01614237,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.6215867041649854,
      "language_loss": 0.81720203,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.83931518,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 2.7453219890594482
    },
    {
      "auxiliary_loss_clip": 0.01196237,
      "auxiliary_loss_mlp": 0.01034802,
      "balance_loss_clip": 0.97745854,
      "balance_loss_mlp": 1.02626681,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.940100661435853,
      "language_loss": 0.856305,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.87861544,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.7258057594299316
    },
    {
      "auxiliary_loss_clip": 0.01200083,
      "auxiliary_loss_mlp": 0.01040207,
      "balance_loss_clip": 1.02086258,
      "balance_loss_mlp": 1.03164792,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.2253989432356307,
      "language_loss": 0.62849492,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65089786,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 2.866410255432129
    },
    {
      "auxiliary_loss_clip": 0.01186073,
      "auxiliary_loss_mlp": 0.01027642,
      "balance_loss_clip": 0.97738385,
      "balance_loss_mlp": 1.01918387,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 1.8715328338513073,
      "language_loss": 0.81509429,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.83723146,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.704596996307373
    },
    {
      "auxiliary_loss_clip": 0.01178733,
      "auxiliary_loss_mlp": 0.01042685,
      "balance_loss_clip": 0.89933693,
      "balance_loss_mlp": 1.03429854,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 2.50489195233826,
      "language_loss": 0.86535215,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88756633,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.7876410484313965
    },
    {
      "auxiliary_loss_clip": 0.0118901,
      "auxiliary_loss_mlp": 0.01027381,
      "balance_loss_clip": 0.93986017,
      "balance_loss_mlp": 1.0189352,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 1.7564421829474461,
      "language_loss": 0.71688867,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.73905265,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 2.770764112472534
    },
    {
      "auxiliary_loss_clip": 0.01181243,
      "auxiliary_loss_mlp": 0.01035303,
      "balance_loss_clip": 0.97772706,
      "balance_loss_mlp": 1.02635074,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 2.240545104412698,
      "language_loss": 0.80689394,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.82905936,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 3.7208714485168457
    },
    {
      "auxiliary_loss_clip": 0.011869,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 0.9770807,
      "balance_loss_mlp": 1.02069736,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 1.7179417655666636,
      "language_loss": 0.67910433,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70126528,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 2.819744825363159
    },
    {
      "auxiliary_loss_clip": 0.01195386,
      "auxiliary_loss_mlp": 0.01032608,
      "balance_loss_clip": 1.05586839,
      "balance_loss_mlp": 1.02293992,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.347657287657787,
      "language_loss": 0.72712344,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.74940336,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 3.659087657928467
    },
    {
      "auxiliary_loss_clip": 0.01191086,
      "auxiliary_loss_mlp": 0.01034265,
      "balance_loss_clip": 0.97911865,
      "balance_loss_mlp": 1.02615881,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 1.9488707362031832,
      "language_loss": 0.87605083,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.89830434,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.7247395515441895
    },
    {
      "auxiliary_loss_clip": 0.01199664,
      "auxiliary_loss_mlp": 0.01034271,
      "balance_loss_clip": 1.05548286,
      "balance_loss_mlp": 1.02478826,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 2.5037195184364833,
      "language_loss": 0.70571762,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.72805703,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.636394500732422
    },
    {
      "auxiliary_loss_clip": 0.01184536,
      "auxiliary_loss_mlp": 0.01028499,
      "balance_loss_clip": 0.93743193,
      "balance_loss_mlp": 1.0199641,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 27.860795477029747,
      "language_loss": 0.86638206,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.88851237,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 3.6602585315704346
    },
    {
      "auxiliary_loss_clip": 0.01174022,
      "auxiliary_loss_mlp": 0.0102885,
      "balance_loss_clip": 0.89736831,
      "balance_loss_mlp": 1.0203917,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 1.800206942164576,
      "language_loss": 0.79498488,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.81701362,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 3.664416790008545
    },
    {
      "auxiliary_loss_clip": 0.01196837,
      "auxiliary_loss_mlp": 0.01032294,
      "balance_loss_clip": 1.05637765,
      "balance_loss_mlp": 1.02384758,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 2.0876145605267094,
      "language_loss": 0.85625398,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87854522,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.6202101707458496
    },
    {
      "auxiliary_loss_clip": 0.01200014,
      "auxiliary_loss_mlp": 0.01028667,
      "balance_loss_clip": 0.94222409,
      "balance_loss_mlp": 1.01951134,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 2.0819049777496574,
      "language_loss": 0.77559131,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.79787815,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.7241883277893066
    },
    {
      "auxiliary_loss_clip": 0.01186867,
      "auxiliary_loss_mlp": 0.01025484,
      "balance_loss_clip": 0.97791725,
      "balance_loss_mlp": 1.01665676,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 2.265500129112915,
      "language_loss": 0.73449546,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75661898,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.7175300121307373
    },
    {
      "auxiliary_loss_clip": 0.0119777,
      "auxiliary_loss_mlp": 0.01032252,
      "balance_loss_clip": 1.05627394,
      "balance_loss_mlp": 1.02346051,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 2.5711876366675264,
      "language_loss": 0.64737844,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.66967869,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.690131425857544
    },
    {
      "auxiliary_loss_clip": 0.01194012,
      "auxiliary_loss_mlp": 0.01030808,
      "balance_loss_clip": 0.97963554,
      "balance_loss_mlp": 1.02230835,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 1.8711979837091615,
      "language_loss": 0.72142637,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74367464,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.689628839492798
    },
    {
      "auxiliary_loss_clip": 0.01196081,
      "auxiliary_loss_mlp": 0.01029556,
      "balance_loss_clip": 1.05649734,
      "balance_loss_mlp": 1.02125287,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 1.8600008279299054,
      "language_loss": 0.73114258,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.75339895,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.6890952587127686
    },
    {
      "auxiliary_loss_clip": 0.01175714,
      "auxiliary_loss_mlp": 0.01124823,
      "balance_loss_clip": 0.97355044,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 2.2502881387318743,
      "language_loss": 0.84417897,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.86718434,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.7564890384674072
    },
    {
      "auxiliary_loss_clip": 0.01195194,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.05665779,
      "balance_loss_mlp": 1.02219272,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 1.7737832486034848,
      "language_loss": 0.69677913,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.71903712,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 2.651423454284668
    },
    {
      "auxiliary_loss_clip": 0.01191034,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.01926661,
      "balance_loss_mlp": 1.02039433,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.6373008505004445,
      "language_loss": 0.84444326,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.866638,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.6597650051116943
    },
    {
      "auxiliary_loss_clip": 0.01192841,
      "auxiliary_loss_mlp": 0.01037688,
      "balance_loss_clip": 1.01942515,
      "balance_loss_mlp": 1.02840781,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 1.830324551749091,
      "language_loss": 0.6823144,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70461965,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.630260467529297
    },
    {
      "auxiliary_loss_clip": 0.0119575,
      "auxiliary_loss_mlp": 0.01038758,
      "balance_loss_clip": 0.98266119,
      "balance_loss_mlp": 1.02908421,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 1.6723859640083363,
      "language_loss": 0.79034376,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81268883,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.717601776123047
    },
    {
      "auxiliary_loss_clip": 0.01201617,
      "auxiliary_loss_mlp": 0.010271,
      "balance_loss_clip": 0.9405511,
      "balance_loss_mlp": 1.01882076,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 1.8296431103720563,
      "language_loss": 0.69944966,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72173679,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 2.7995052337646484
    },
    {
      "auxiliary_loss_clip": 0.01188486,
      "auxiliary_loss_mlp": 0.01030366,
      "balance_loss_clip": 0.86556649,
      "balance_loss_mlp": 1.02198601,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 1.827826533877319,
      "language_loss": 0.89376062,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.91594917,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 2.845081090927124
    },
    {
      "auxiliary_loss_clip": 0.0119709,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 0.94074583,
      "balance_loss_mlp": 1.02242172,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 1.8297785750632403,
      "language_loss": 0.79910648,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.82138759,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 2.7776038646698
    },
    {
      "auxiliary_loss_clip": 0.01171429,
      "auxiliary_loss_mlp": 0.01032185,
      "balance_loss_clip": 0.93677044,
      "balance_loss_mlp": 1.02360225,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 2.023604483769012,
      "language_loss": 0.85142469,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87346083,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.7661943435668945
    },
    {
      "auxiliary_loss_clip": 0.01190975,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.0205915,
      "balance_loss_mlp": 1.01947594,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 1.8774269638307994,
      "language_loss": 0.83893764,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86112523,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.713223457336426
    },
    {
      "auxiliary_loss_clip": 0.01178743,
      "auxiliary_loss_mlp": 0.01039105,
      "balance_loss_clip": 0.90122402,
      "balance_loss_mlp": 1.03060532,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.1108103938371316,
      "language_loss": 0.79037535,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.81255388,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 2.862790584564209
    },
    {
      "auxiliary_loss_clip": 0.01194059,
      "auxiliary_loss_mlp": 0.01033842,
      "balance_loss_clip": 1.01615119,
      "balance_loss_mlp": 1.0247587,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 2.077930071611614,
      "language_loss": 0.7696659,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79194486,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.7032461166381836
    },
    {
      "auxiliary_loss_clip": 0.01191065,
      "auxiliary_loss_mlp": 0.01031167,
      "balance_loss_clip": 0.97984254,
      "balance_loss_mlp": 1.02270865,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 1.9960415170083423,
      "language_loss": 0.809425,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.8316474,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 3.726520538330078
    },
    {
      "auxiliary_loss_clip": 0.01186469,
      "auxiliary_loss_mlp": 0.01033285,
      "balance_loss_clip": 0.97955668,
      "balance_loss_mlp": 1.02525592,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 1.6470294684620872,
      "language_loss": 0.68214166,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70433915,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 2.728177309036255
    },
    {
      "auxiliary_loss_clip": 0.01198197,
      "auxiliary_loss_mlp": 0.01031541,
      "balance_loss_clip": 1.0158391,
      "balance_loss_mlp": 1.02291584,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 3.9700230289449236,
      "language_loss": 0.82578331,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.84808064,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.6585888862609863
    },
    {
      "auxiliary_loss_clip": 0.01190594,
      "auxiliary_loss_mlp": 0.01035382,
      "balance_loss_clip": 1.05559516,
      "balance_loss_mlp": 1.0277586,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 1.5917449757391176,
      "language_loss": 0.88260227,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90486205,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 3.647597551345825
    },
    {
      "auxiliary_loss_clip": 0.01191201,
      "auxiliary_loss_mlp": 0.01124933,
      "balance_loss_clip": 0.94216704,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 1.538527042566895,
      "language_loss": 0.85678542,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.87994671,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.8510684967041016
    },
    {
      "auxiliary_loss_clip": 0.01178668,
      "auxiliary_loss_mlp": 0.01027115,
      "balance_loss_clip": 0.89888078,
      "balance_loss_mlp": 1.0193603,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.5765620738440287,
      "language_loss": 0.52434999,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54640782,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.721595287322998
    },
    {
      "auxiliary_loss_clip": 0.01191261,
      "auxiliary_loss_mlp": 0.01036936,
      "balance_loss_clip": 1.05413318,
      "balance_loss_mlp": 1.0278641,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 2.5340285221495606,
      "language_loss": 0.83665502,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.85893703,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 3.618786334991455
    },
    {
      "auxiliary_loss_clip": 0.01195361,
      "auxiliary_loss_mlp": 0.01031971,
      "balance_loss_clip": 1.01779842,
      "balance_loss_mlp": 1.0225718,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 2.3689147618940076,
      "language_loss": 0.90417302,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.92644632,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 3.6073877811431885
    },
    {
      "auxiliary_loss_clip": 0.01187105,
      "auxiliary_loss_mlp": 0.01031526,
      "balance_loss_clip": 0.97773111,
      "balance_loss_mlp": 1.02262068,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 2.819840932391084,
      "language_loss": 0.85900962,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.88119596,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.5866641998291016
    },
    {
      "auxiliary_loss_clip": 0.01194095,
      "auxiliary_loss_mlp": 0.01030763,
      "balance_loss_clip": 1.01596606,
      "balance_loss_mlp": 1.02287161,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 1.7488363199120198,
      "language_loss": 0.81124473,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.83349335,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.7663450241088867
    },
    {
      "auxiliary_loss_clip": 0.01188156,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 0.97930807,
      "balance_loss_mlp": 1.0226779,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 1.6396577563294197,
      "language_loss": 0.79864407,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.82083774,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.705282688140869
    },
    {
      "auxiliary_loss_clip": 0.01195749,
      "auxiliary_loss_mlp": 0.01035919,
      "balance_loss_clip": 1.05831456,
      "balance_loss_mlp": 1.02808666,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 1.7236361524728019,
      "language_loss": 0.75995266,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78226936,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.726907968521118
    },
    {
      "auxiliary_loss_clip": 0.01200348,
      "auxiliary_loss_mlp": 0.01030997,
      "balance_loss_clip": 0.98199749,
      "balance_loss_mlp": 1.02259862,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.5325765470721717,
      "language_loss": 0.73180532,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.7541188,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.680854558944702
    },
    {
      "auxiliary_loss_clip": 0.01189688,
      "auxiliary_loss_mlp": 0.01032548,
      "balance_loss_clip": 0.98011947,
      "balance_loss_mlp": 1.02383399,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 1.7559281770826638,
      "language_loss": 0.70707333,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.72929573,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.8185372352600098
    },
    {
      "auxiliary_loss_clip": 0.01193745,
      "auxiliary_loss_mlp": 0.01029708,
      "balance_loss_clip": 0.94227707,
      "balance_loss_mlp": 1.02108932,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.663979432041294,
      "language_loss": 0.85710007,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.87933463,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.7267234325408936
    },
    {
      "auxiliary_loss_clip": 0.01195131,
      "auxiliary_loss_mlp": 0.01033257,
      "balance_loss_clip": 1.05649006,
      "balance_loss_mlp": 1.02498412,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 1.7560597835977487,
      "language_loss": 0.79976416,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82204801,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 2.638112783432007
    },
    {
      "auxiliary_loss_clip": 0.01193078,
      "auxiliary_loss_mlp": 0.01125129,
      "balance_loss_clip": 1.05549157,
      "balance_loss_mlp": 0.0,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.87020294789648,
      "language_loss": 0.83067703,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.85385907,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.6631596088409424
    },
    {
      "auxiliary_loss_clip": 0.01193007,
      "auxiliary_loss_mlp": 0.01033756,
      "balance_loss_clip": 0.97669202,
      "balance_loss_mlp": 1.02496409,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 2.038322720168869,
      "language_loss": 0.76410568,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.78637326,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.730715274810791
    },
    {
      "auxiliary_loss_clip": 0.0119183,
      "auxiliary_loss_mlp": 0.01039523,
      "balance_loss_clip": 1.01487672,
      "balance_loss_mlp": 1.03117847,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 1.9075699912298736,
      "language_loss": 0.86934918,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.89166272,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.69920015335083
    },
    {
      "auxiliary_loss_clip": 0.01193071,
      "auxiliary_loss_mlp": 0.01031929,
      "balance_loss_clip": 1.05481005,
      "balance_loss_mlp": 1.02310133,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 2.269733580136174,
      "language_loss": 0.72897989,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.75122988,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 2.649411678314209
    },
    {
      "auxiliary_loss_clip": 0.01191377,
      "auxiliary_loss_mlp": 0.01025643,
      "balance_loss_clip": 0.97911012,
      "balance_loss_mlp": 1.0175488,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 1.7756371612407833,
      "language_loss": 0.8521204,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87429059,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 2.7440600395202637
    },
    {
      "auxiliary_loss_clip": 0.01191574,
      "auxiliary_loss_mlp": 0.01030261,
      "balance_loss_clip": 1.01604724,
      "balance_loss_mlp": 1.02220845,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 2.2939707732302703,
      "language_loss": 0.89039636,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91261476,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 2.768230676651001
    },
    {
      "auxiliary_loss_clip": 0.01176899,
      "auxiliary_loss_mlp": 0.01125289,
      "balance_loss_clip": 0.97442669,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 1.9459178367553482,
      "language_loss": 0.66489363,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.6879155,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.8422021865844727
    },
    {
      "auxiliary_loss_clip": 0.01193305,
      "auxiliary_loss_mlp": 0.01029913,
      "balance_loss_clip": 0.9821955,
      "balance_loss_mlp": 1.021837,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 1.4941700720378899,
      "language_loss": 0.77401519,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79624736,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.7280635833740234
    },
    {
      "auxiliary_loss_clip": 0.01192385,
      "auxiliary_loss_mlp": 0.01033797,
      "balance_loss_clip": 1.05451024,
      "balance_loss_mlp": 1.02564883,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 2.057621046082418,
      "language_loss": 0.74226373,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.76452553,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 2.6619057655334473
    },
    {
      "auxiliary_loss_clip": 0.01176747,
      "auxiliary_loss_mlp": 0.01025133,
      "balance_loss_clip": 0.93870854,
      "balance_loss_mlp": 1.0168128,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 1.7110278061997566,
      "language_loss": 0.72136939,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74338824,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 2.7272422313690186
    },
    {
      "auxiliary_loss_clip": 0.01195978,
      "auxiliary_loss_mlp": 0.01125459,
      "balance_loss_clip": 0.98027188,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 1.8989437654829076,
      "language_loss": 0.89013541,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.91334981,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.7409791946411133
    },
    {
      "auxiliary_loss_clip": 0.01192416,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.01662588,
      "balance_loss_mlp": 1.01907659,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 2.0280707489634797,
      "language_loss": 0.80188489,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82408804,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 3.752699375152588
    },
    {
      "auxiliary_loss_clip": 0.01185442,
      "auxiliary_loss_mlp": 0.01036326,
      "balance_loss_clip": 0.98072362,
      "balance_loss_mlp": 1.02838051,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 1.880625789850816,
      "language_loss": 0.78445655,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80667424,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 2.7245404720306396
    },
    {
      "auxiliary_loss_clip": 0.01186612,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.01536727,
      "balance_loss_mlp": 1.01840067,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 2.0103783919400575,
      "language_loss": 0.70426786,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72640789,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 3.738264799118042
    },
    {
      "auxiliary_loss_clip": 0.01194599,
      "auxiliary_loss_mlp": 0.01030751,
      "balance_loss_clip": 0.97694916,
      "balance_loss_mlp": 1.02192342,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 1.8524105183878696,
      "language_loss": 0.92117739,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.9434309,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.7585699558258057
    },
    {
      "auxiliary_loss_clip": 0.01191332,
      "auxiliary_loss_mlp": 0.01036319,
      "balance_loss_clip": 1.01319528,
      "balance_loss_mlp": 1.02730072,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 2.8120645194629827,
      "language_loss": 0.77708733,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.79936397,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.677809476852417
    },
    {
      "auxiliary_loss_clip": 0.01181721,
      "auxiliary_loss_mlp": 0.01124523,
      "balance_loss_clip": 0.97454709,
      "balance_loss_mlp": 0.0,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.7247683764387935,
      "language_loss": 0.64828324,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.67134571,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 3.6786882877349854
    },
    {
      "auxiliary_loss_clip": 0.01191656,
      "auxiliary_loss_mlp": 0.01033869,
      "balance_loss_clip": 1.01708317,
      "balance_loss_mlp": 1.02432013,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.207942554798071,
      "language_loss": 0.71000636,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.7322616,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 3.624160051345825
    },
    {
      "auxiliary_loss_clip": 0.01195868,
      "auxiliary_loss_mlp": 0.01032185,
      "balance_loss_clip": 1.01959419,
      "balance_loss_mlp": 1.02353644,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 3.5422217214493728,
      "language_loss": 0.67455065,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.69683123,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.5932648181915283
    },
    {
      "auxiliary_loss_clip": 0.01191846,
      "auxiliary_loss_mlp": 0.0103467,
      "balance_loss_clip": 0.9796856,
      "balance_loss_mlp": 1.02599764,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 1.9979867187438807,
      "language_loss": 0.65443289,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67669797,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.870617151260376
    },
    {
      "auxiliary_loss_clip": 0.01182728,
      "auxiliary_loss_mlp": 0.01031882,
      "balance_loss_clip": 0.93525732,
      "balance_loss_mlp": 1.02251232,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.343810485021572,
      "language_loss": 0.65404391,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67619002,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.8361058235168457
    },
    {
      "auxiliary_loss_clip": 0.01191486,
      "auxiliary_loss_mlp": 0.0103435,
      "balance_loss_clip": 1.05494857,
      "balance_loss_mlp": 1.0254333,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.6085648227166107,
      "language_loss": 0.79952049,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82177877,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.71669602394104
    },
    {
      "auxiliary_loss_clip": 0.01094479,
      "auxiliary_loss_mlp": 0.01004615,
      "balance_loss_clip": 0.87264818,
      "balance_loss_mlp": 1.00231445,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.808670683492874,
      "language_loss": 0.6024009,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.62339181,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.34729266166687
    },
    {
      "auxiliary_loss_clip": 0.01182873,
      "auxiliary_loss_mlp": 0.01029149,
      "balance_loss_clip": 0.94134581,
      "balance_loss_mlp": 1.02156687,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 1.7389843374829865,
      "language_loss": 0.77215046,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.79427069,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.7815473079681396
    },
    {
      "auxiliary_loss_clip": 0.01195443,
      "auxiliary_loss_mlp": 0.01031682,
      "balance_loss_clip": 1.05717659,
      "balance_loss_mlp": 1.02311039,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 1.739472796573222,
      "language_loss": 0.78983349,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.8121047,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.7012593746185303
    },
    {
      "auxiliary_loss_clip": 0.01189561,
      "auxiliary_loss_mlp": 0.01036331,
      "balance_loss_clip": 0.93800545,
      "balance_loss_mlp": 1.02753329,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 1.7425517061629576,
      "language_loss": 0.75554466,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.7778036,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.7752890586853027
    },
    {
      "auxiliary_loss_clip": 0.01190836,
      "auxiliary_loss_mlp": 0.0103119,
      "balance_loss_clip": 1.01385462,
      "balance_loss_mlp": 1.02255642,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 1.710771804121954,
      "language_loss": 0.68760514,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.7098254,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 2.7427945137023926
    },
    {
      "auxiliary_loss_clip": 0.01173696,
      "auxiliary_loss_mlp": 0.0103467,
      "balance_loss_clip": 0.82297695,
      "balance_loss_mlp": 1.02568734,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 1.9699201402995807,
      "language_loss": 0.66949981,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.69158351,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.8205769062042236
    },
    {
      "auxiliary_loss_clip": 0.01182625,
      "auxiliary_loss_mlp": 0.01031486,
      "balance_loss_clip": 0.9794656,
      "balance_loss_mlp": 1.02277207,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.4419404186935787,
      "language_loss": 0.74394631,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.76608741,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.913639783859253
    },
    {
      "auxiliary_loss_clip": 0.01179187,
      "auxiliary_loss_mlp": 0.01025252,
      "balance_loss_clip": 1.0125159,
      "balance_loss_mlp": 1.01688385,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.73346942449734,
      "language_loss": 0.73910713,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.76115143,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 2.725569009780884
    },
    {
      "auxiliary_loss_clip": 0.01189355,
      "auxiliary_loss_mlp": 0.01030391,
      "balance_loss_clip": 0.97841042,
      "balance_loss_mlp": 1.02162886,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 1.8373097089084762,
      "language_loss": 0.7352888,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75748628,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 2.7055506706237793
    },
    {
      "auxiliary_loss_clip": 0.01190483,
      "auxiliary_loss_mlp": 0.0102981,
      "balance_loss_clip": 1.05339026,
      "balance_loss_mlp": 1.02147114,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 2.8329104727897065,
      "language_loss": 0.75899339,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.78119636,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.6683900356292725
    },
    {
      "auxiliary_loss_clip": 0.01190951,
      "auxiliary_loss_mlp": 0.01033136,
      "balance_loss_clip": 1.01523733,
      "balance_loss_mlp": 1.02437365,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 1.963013109894051,
      "language_loss": 0.7919395,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81418037,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 2.6125426292419434
    },
    {
      "auxiliary_loss_clip": 0.01192325,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 0.9402349,
      "balance_loss_mlp": 1.02139175,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 2.15630861575301,
      "language_loss": 0.71296203,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73518538,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.7245888710021973
    },
    {
      "auxiliary_loss_clip": 0.01182085,
      "auxiliary_loss_mlp": 0.01033988,
      "balance_loss_clip": 0.97604597,
      "balance_loss_mlp": 1.02537537,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.790019392231014,
      "language_loss": 0.8250283,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.84718901,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.7770674228668213
    },
    {
      "auxiliary_loss_clip": 0.01180929,
      "auxiliary_loss_mlp": 0.01032165,
      "balance_loss_clip": 0.97742522,
      "balance_loss_mlp": 1.02415419,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.645629043288728,
      "language_loss": 0.6164726,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.63860357,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 2.7918365001678467
    },
    {
      "auxiliary_loss_clip": 0.01187311,
      "auxiliary_loss_mlp": 0.01124797,
      "balance_loss_clip": 0.93573964,
      "balance_loss_mlp": 0.0,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 1.6987572925884438,
      "language_loss": 0.74880242,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.77192348,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.7908246517181396
    },
    {
      "auxiliary_loss_clip": 0.01185757,
      "auxiliary_loss_mlp": 0.01030181,
      "balance_loss_clip": 0.97827303,
      "balance_loss_mlp": 1.02216387,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 1.7947223030317419,
      "language_loss": 0.72783905,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.74999845,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 3.7886574268341064
    },
    {
      "auxiliary_loss_clip": 0.011817,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 0.93525219,
      "balance_loss_mlp": 1.02057648,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.622305591067206,
      "language_loss": 0.70466375,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.72677445,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 2.765206813812256
    },
    {
      "auxiliary_loss_clip": 0.01190275,
      "auxiliary_loss_mlp": 0.01124925,
      "balance_loss_clip": 0.94005102,
      "balance_loss_mlp": 0.0,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 1.9071829130202629,
      "language_loss": 0.65682602,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.67997801,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 3.7343966960906982
    },
    {
      "auxiliary_loss_clip": 0.01188342,
      "auxiliary_loss_mlp": 0.01029559,
      "balance_loss_clip": 0.98014647,
      "balance_loss_mlp": 1.02124429,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 1.7592567709328246,
      "language_loss": 0.80754387,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.829723,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.7536327838897705
    },
    {
      "auxiliary_loss_clip": 0.01182589,
      "auxiliary_loss_mlp": 0.01033334,
      "balance_loss_clip": 1.01605952,
      "balance_loss_mlp": 1.02532303,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.7336590891921337,
      "language_loss": 0.74647671,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.76863587,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.6317756175994873
    },
    {
      "auxiliary_loss_clip": 0.01197428,
      "auxiliary_loss_mlp": 0.01029773,
      "balance_loss_clip": 0.97844619,
      "balance_loss_mlp": 1.02090406,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 1.609623635693904,
      "language_loss": 0.78798091,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.81025302,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 3.736506700515747
    },
    {
      "auxiliary_loss_clip": 0.01190651,
      "auxiliary_loss_mlp": 0.01032893,
      "balance_loss_clip": 1.01960039,
      "balance_loss_mlp": 1.02509654,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 1.885275476766569,
      "language_loss": 0.78469992,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80693537,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.6204121112823486
    },
    {
      "auxiliary_loss_clip": 0.01200186,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 0.98210853,
      "balance_loss_mlp": 1.02525973,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 1.4988087828745946,
      "language_loss": 0.70670664,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.7290439,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 3.6098389625549316
    },
    {
      "auxiliary_loss_clip": 0.01180091,
      "auxiliary_loss_mlp": 0.01031183,
      "balance_loss_clip": 0.97553617,
      "balance_loss_mlp": 1.02202785,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 1.9098225356848268,
      "language_loss": 0.70486426,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72697699,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.6875314712524414
    },
    {
      "auxiliary_loss_clip": 0.01198368,
      "auxiliary_loss_mlp": 0.01033183,
      "balance_loss_clip": 1.01890707,
      "balance_loss_mlp": 1.02428997,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 1.7528416447675246,
      "language_loss": 0.78112686,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80344242,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.670952796936035
    },
    {
      "auxiliary_loss_clip": 0.0118638,
      "auxiliary_loss_mlp": 0.01035028,
      "balance_loss_clip": 0.97793138,
      "balance_loss_mlp": 1.025962,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 2.458798305504796,
      "language_loss": 0.84263229,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86484641,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.698072910308838
    },
    {
      "auxiliary_loss_clip": 0.01177127,
      "auxiliary_loss_mlp": 0.01031552,
      "balance_loss_clip": 0.93612158,
      "balance_loss_mlp": 1.02320755,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.5613773096458692,
      "language_loss": 0.75583607,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.77792287,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.7644801139831543
    },
    {
      "auxiliary_loss_clip": 0.01177851,
      "auxiliary_loss_mlp": 0.01036292,
      "balance_loss_clip": 0.89846063,
      "balance_loss_mlp": 1.02773261,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 1.8146087874305605,
      "language_loss": 0.89185101,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.9139924,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.84908127784729
    },
    {
      "auxiliary_loss_clip": 0.01188422,
      "auxiliary_loss_mlp": 0.0103719,
      "balance_loss_clip": 0.94136405,
      "balance_loss_mlp": 1.02822518,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 2.53256553292484,
      "language_loss": 0.6789788,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.70123482,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.7951271533966064
    },
    {
      "auxiliary_loss_clip": 0.01195991,
      "auxiliary_loss_mlp": 0.01036583,
      "balance_loss_clip": 1.05733252,
      "balance_loss_mlp": 1.02781534,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 1.7750053401691834,
      "language_loss": 0.80230319,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82462895,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.6320478916168213
    },
    {
      "auxiliary_loss_clip": 0.01188882,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 1.01875257,
      "balance_loss_mlp": 1.01873517,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 2.5593591436404313,
      "language_loss": 0.76711375,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.7892729,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 2.648137092590332
    },
    {
      "auxiliary_loss_clip": 0.01169309,
      "auxiliary_loss_mlp": 0.01039505,
      "balance_loss_clip": 0.86140466,
      "balance_loss_mlp": 1.0302608,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 3.1064172387275146,
      "language_loss": 0.77136707,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.79345524,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.8084819316864014
    },
    {
      "auxiliary_loss_clip": 0.01189674,
      "auxiliary_loss_mlp": 0.01032075,
      "balance_loss_clip": 0.94367683,
      "balance_loss_mlp": 1.02449894,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 1.8153344990217986,
      "language_loss": 0.84210706,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86432457,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.6852328777313232
    },
    {
      "auxiliary_loss_clip": 0.01180543,
      "auxiliary_loss_mlp": 0.01029512,
      "balance_loss_clip": 0.97666013,
      "balance_loss_mlp": 1.02077961,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 2.399187604198925,
      "language_loss": 0.62257099,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.6446715,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.799711227416992
    },
    {
      "auxiliary_loss_clip": 0.0118949,
      "auxiliary_loss_mlp": 0.01035134,
      "balance_loss_clip": 0.97941434,
      "balance_loss_mlp": 1.02535248,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 2.4881347606537876,
      "language_loss": 0.70038855,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.72263479,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 2.7321197986602783
    },
    {
      "auxiliary_loss_clip": 0.01199442,
      "auxiliary_loss_mlp": 0.01034662,
      "balance_loss_clip": 1.0177114,
      "balance_loss_mlp": 1.02581096,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 2.4582376328478497,
      "language_loss": 0.83009481,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.85243589,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 2.7051820755004883
    },
    {
      "auxiliary_loss_clip": 0.01188561,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 0.97961253,
      "balance_loss_mlp": 1.01845646,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 1.6268562553010364,
      "language_loss": 0.73864496,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.76079166,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.72733736038208
    },
    {
      "auxiliary_loss_clip": 0.01184163,
      "auxiliary_loss_mlp": 0.01040741,
      "balance_loss_clip": 1.01575959,
      "balance_loss_mlp": 1.03259289,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 2.000895975613671,
      "language_loss": 0.77639508,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79864419,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.676266670227051
    },
    {
      "auxiliary_loss_clip": 0.0118828,
      "auxiliary_loss_mlp": 0.01030312,
      "balance_loss_clip": 1.01850533,
      "balance_loss_mlp": 1.02224708,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 2.2163317122984294,
      "language_loss": 0.88165349,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90383941,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.7197325229644775
    },
    {
      "auxiliary_loss_clip": 0.01176054,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 0.90081042,
      "balance_loss_mlp": 1.01751792,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 2.710243606612998,
      "language_loss": 0.6050328,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62705773,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 2.910104990005493
    },
    {
      "auxiliary_loss_clip": 0.01180053,
      "auxiliary_loss_mlp": 0.0103207,
      "balance_loss_clip": 0.93741071,
      "balance_loss_mlp": 1.02327871,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 1.7566477635141584,
      "language_loss": 0.77246332,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.79458463,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 2.7559571266174316
    },
    {
      "auxiliary_loss_clip": 0.01178025,
      "auxiliary_loss_mlp": 0.01032406,
      "balance_loss_clip": 0.8985433,
      "balance_loss_mlp": 1.02365029,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 1.8292363553982485,
      "language_loss": 0.67160439,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69370878,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 3.88474440574646
    },
    {
      "auxiliary_loss_clip": 0.01183285,
      "auxiliary_loss_mlp": 0.01124756,
      "balance_loss_clip": 0.97537625,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 1.8564303895772631,
      "language_loss": 0.84068954,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.86376995,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 2.714338779449463
    },
    {
      "auxiliary_loss_clip": 0.0119232,
      "auxiliary_loss_mlp": 0.01032318,
      "balance_loss_clip": 1.02058232,
      "balance_loss_mlp": 1.02379417,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 2.146256465928803,
      "language_loss": 0.82451135,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.84675771,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 3.7269952297210693
    },
    {
      "auxiliary_loss_clip": 0.0118771,
      "auxiliary_loss_mlp": 0.01031529,
      "balance_loss_clip": 1.01839578,
      "balance_loss_mlp": 1.02342844,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 1.737051543442168,
      "language_loss": 0.78536654,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.80755895,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.7283761501312256
    },
    {
      "auxiliary_loss_clip": 0.01180462,
      "auxiliary_loss_mlp": 0.01031825,
      "balance_loss_clip": 0.93997514,
      "balance_loss_mlp": 1.02367687,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 1.9334141523138857,
      "language_loss": 0.86595333,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.88807619,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.73494029045105
    },
    {
      "auxiliary_loss_clip": 0.01187633,
      "auxiliary_loss_mlp": 0.01032667,
      "balance_loss_clip": 0.9439733,
      "balance_loss_mlp": 1.02434063,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 1.8865114956218836,
      "language_loss": 0.67479724,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.69700027,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 3.763049840927124
    },
    {
      "auxiliary_loss_clip": 0.01192041,
      "auxiliary_loss_mlp": 0.01027733,
      "balance_loss_clip": 1.02069926,
      "balance_loss_mlp": 1.01948369,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 1.976670176631696,
      "language_loss": 0.71514535,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73734307,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.6552491188049316
    },
    {
      "auxiliary_loss_clip": 0.01185601,
      "auxiliary_loss_mlp": 0.01034639,
      "balance_loss_clip": 1.01735544,
      "balance_loss_mlp": 1.02607989,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 1.9821138704931023,
      "language_loss": 0.75940228,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.78160477,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 3.600180149078369
    },
    {
      "auxiliary_loss_clip": 0.01199977,
      "auxiliary_loss_mlp": 0.01032051,
      "balance_loss_clip": 1.06111479,
      "balance_loss_mlp": 1.02331853,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 1.5748510836492629,
      "language_loss": 0.82145953,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.8437798,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.6471962928771973
    },
    {
      "auxiliary_loss_clip": 0.01200483,
      "auxiliary_loss_mlp": 0.01032319,
      "balance_loss_clip": 0.98191047,
      "balance_loss_mlp": 1.02372408,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 1.8571929035565276,
      "language_loss": 0.85180974,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.87413776,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.7474863529205322
    },
    {
      "auxiliary_loss_clip": 0.0118778,
      "auxiliary_loss_mlp": 0.01029997,
      "balance_loss_clip": 0.98191494,
      "balance_loss_mlp": 1.02207565,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 2.025393527342934,
      "language_loss": 0.7918424,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.81402016,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.7694313526153564
    },
    {
      "auxiliary_loss_clip": 0.01194147,
      "auxiliary_loss_mlp": 0.01031459,
      "balance_loss_clip": 0.98053336,
      "balance_loss_mlp": 1.02299511,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 1.7026154155836266,
      "language_loss": 0.80446655,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.82672262,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.6517586708068848
    },
    {
      "auxiliary_loss_clip": 0.01189503,
      "auxiliary_loss_mlp": 0.01027426,
      "balance_loss_clip": 0.98094308,
      "balance_loss_mlp": 1.01862884,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 2.058356437950254,
      "language_loss": 0.72505486,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.74722409,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.7598323822021484
    },
    {
      "auxiliary_loss_clip": 0.01189421,
      "auxiliary_loss_mlp": 0.01033144,
      "balance_loss_clip": 0.86550403,
      "balance_loss_mlp": 1.02466798,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 2.0089877297984406,
      "language_loss": 0.72075987,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.74298555,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.8106751441955566
    },
    {
      "auxiliary_loss_clip": 0.01189889,
      "auxiliary_loss_mlp": 0.0102746,
      "balance_loss_clip": 0.94028205,
      "balance_loss_mlp": 1.01844823,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 1.8744840760266814,
      "language_loss": 0.7567665,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.77893996,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.776237964630127
    },
    {
      "auxiliary_loss_clip": 0.01193739,
      "auxiliary_loss_mlp": 0.01033062,
      "balance_loss_clip": 1.01891398,
      "balance_loss_mlp": 1.02417529,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 2.1410577525452092,
      "language_loss": 0.94699192,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.96925986,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.6915552616119385
    },
    {
      "auxiliary_loss_clip": 0.01194687,
      "auxiliary_loss_mlp": 0.01027791,
      "balance_loss_clip": 0.98098886,
      "balance_loss_mlp": 1.01907659,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 1.993277395295914,
      "language_loss": 0.81383705,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83606178,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.606861114501953
    },
    {
      "auxiliary_loss_clip": 0.01184267,
      "auxiliary_loss_mlp": 0.01027217,
      "balance_loss_clip": 0.93936521,
      "balance_loss_mlp": 1.01871109,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 1.7511853478392008,
      "language_loss": 0.77260953,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.79472435,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.715524911880493
    },
    {
      "auxiliary_loss_clip": 0.01197784,
      "auxiliary_loss_mlp": 0.01024479,
      "balance_loss_clip": 0.98175597,
      "balance_loss_mlp": 1.01623547,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.3491136104010106,
      "language_loss": 0.79895616,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82117873,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.7215676307678223
    },
    {
      "auxiliary_loss_clip": 0.01190331,
      "auxiliary_loss_mlp": 0.01027586,
      "balance_loss_clip": 0.98242694,
      "balance_loss_mlp": 1.01903248,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 1.7814952582754384,
      "language_loss": 0.74155647,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.76373565,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 2.780669927597046
    },
    {
      "auxiliary_loss_clip": 0.01199811,
      "auxiliary_loss_mlp": 0.01031478,
      "balance_loss_clip": 1.05868077,
      "balance_loss_mlp": 1.02235842,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 1.9366773024145534,
      "language_loss": 0.71614468,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.73845756,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 2.641798734664917
    },
    {
      "auxiliary_loss_clip": 0.01197414,
      "auxiliary_loss_mlp": 0.01034762,
      "balance_loss_clip": 1.0208571,
      "balance_loss_mlp": 1.02625036,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 2.397472983705064,
      "language_loss": 0.78634524,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80866694,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 2.6865434646606445
    },
    {
      "auxiliary_loss_clip": 0.01192277,
      "auxiliary_loss_mlp": 0.0103088,
      "balance_loss_clip": 0.97937083,
      "balance_loss_mlp": 1.02106273,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 1.9621970208713873,
      "language_loss": 0.81123698,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.83346856,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.6833415031433105
    },
    {
      "auxiliary_loss_clip": 0.01199768,
      "auxiliary_loss_mlp": 0.01033672,
      "balance_loss_clip": 0.98445636,
      "balance_loss_mlp": 1.02420664,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 1.861680652617834,
      "language_loss": 0.8193481,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.84168249,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.689523935317993
    },
    {
      "auxiliary_loss_clip": 0.01193428,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.01773119,
      "balance_loss_mlp": 1.02114391,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 1.8945358808106005,
      "language_loss": 0.72878218,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.7510128,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 2.7070865631103516
    },
    {
      "auxiliary_loss_clip": 0.01194093,
      "auxiliary_loss_mlp": 0.0112546,
      "balance_loss_clip": 0.94255847,
      "balance_loss_mlp": 0.0,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.5944273106649576,
      "language_loss": 0.80471665,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82791221,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 2.7696070671081543
    },
    {
      "auxiliary_loss_clip": 0.01196985,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 1.02161992,
      "balance_loss_mlp": 1.02321696,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 5.903824093328677,
      "language_loss": 0.77199233,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.79428178,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 3.6566646099090576
    },
    {
      "auxiliary_loss_clip": 0.01179965,
      "auxiliary_loss_mlp": 0.01034069,
      "balance_loss_clip": 0.8997696,
      "balance_loss_mlp": 1.02489555,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 2.5527953173348124,
      "language_loss": 0.81357813,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83571845,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.7071592807769775
    },
    {
      "auxiliary_loss_clip": 0.01195327,
      "auxiliary_loss_mlp": 0.01031454,
      "balance_loss_clip": 1.01922393,
      "balance_loss_mlp": 1.02238822,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 1.8346881476618184,
      "language_loss": 0.77600968,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.7982775,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 3.6917805671691895
    },
    {
      "auxiliary_loss_clip": 0.01203394,
      "auxiliary_loss_mlp": 0.01033528,
      "balance_loss_clip": 1.02179909,
      "balance_loss_mlp": 1.02518344,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 1.3993852475733077,
      "language_loss": 0.68899173,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71136093,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.7105045318603516
    },
    {
      "auxiliary_loss_clip": 0.01190759,
      "auxiliary_loss_mlp": 0.0103075,
      "balance_loss_clip": 1.01908267,
      "balance_loss_mlp": 1.0218327,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 1.8004693422804388,
      "language_loss": 0.69733763,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.7195527,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.763500690460205
    },
    {
      "auxiliary_loss_clip": 0.01104608,
      "auxiliary_loss_mlp": 0.01004171,
      "balance_loss_clip": 1.03546703,
      "balance_loss_mlp": 1.00172734,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0658456632544175,
      "language_loss": 0.64772034,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66880816,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 4.023785591125488
    },
    {
      "auxiliary_loss_clip": 0.01186158,
      "auxiliary_loss_mlp": 0.01032141,
      "balance_loss_clip": 0.97598422,
      "balance_loss_mlp": 1.02399302,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.5077575481557814,
      "language_loss": 0.92768323,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.94986629,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.795625686645508
    },
    {
      "auxiliary_loss_clip": 0.01198347,
      "auxiliary_loss_mlp": 0.01034204,
      "balance_loss_clip": 1.02088165,
      "balance_loss_mlp": 1.02545965,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.7161567350075209,
      "language_loss": 0.69267935,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.71500486,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 3.6577138900756836
    },
    {
      "auxiliary_loss_clip": 0.01188629,
      "auxiliary_loss_mlp": 0.01029551,
      "balance_loss_clip": 0.90156162,
      "balance_loss_mlp": 1.02094436,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 2.6839703681057556,
      "language_loss": 0.80813903,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.83032084,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.8190789222717285
    },
    {
      "auxiliary_loss_clip": 0.01190507,
      "auxiliary_loss_mlp": 0.0103394,
      "balance_loss_clip": 1.01673412,
      "balance_loss_mlp": 1.02451658,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 1.7523565011795224,
      "language_loss": 0.79844552,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.82069004,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.6695330142974854
    },
    {
      "auxiliary_loss_clip": 0.01196034,
      "auxiliary_loss_mlp": 0.01027251,
      "balance_loss_clip": 1.01939154,
      "balance_loss_mlp": 1.01839364,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 1.5967259542996386,
      "language_loss": 0.70825517,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.73048806,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.7632429599761963
    },
    {
      "auxiliary_loss_clip": 0.01192107,
      "auxiliary_loss_mlp": 0.01125811,
      "balance_loss_clip": 0.98067743,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 1.8712839548820945,
      "language_loss": 0.82268488,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84586406,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.6838033199310303
    },
    {
      "auxiliary_loss_clip": 0.01105566,
      "auxiliary_loss_mlp": 0.01002874,
      "balance_loss_clip": 1.00068998,
      "balance_loss_mlp": 1.00037086,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 2.2499433406247924,
      "language_loss": 0.63681817,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.65790254,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.3208224773406982
    },
    {
      "auxiliary_loss_clip": 0.01197101,
      "auxiliary_loss_mlp": 0.01032996,
      "balance_loss_clip": 1.02360797,
      "balance_loss_mlp": 1.02476144,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 1.862883962031307,
      "language_loss": 0.83516216,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.85746306,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.7494986057281494
    },
    {
      "auxiliary_loss_clip": 0.01195083,
      "auxiliary_loss_mlp": 0.01035397,
      "balance_loss_clip": 0.94231582,
      "balance_loss_mlp": 1.02576506,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 1.7431243809051458,
      "language_loss": 0.82883197,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.8511368,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.751988172531128
    },
    {
      "auxiliary_loss_clip": 0.01191583,
      "auxiliary_loss_mlp": 0.01039831,
      "balance_loss_clip": 0.90476632,
      "balance_loss_mlp": 1.03076482,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 2.2840275493393736,
      "language_loss": 0.7838369,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.80615103,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.8275411128997803
    },
    {
      "auxiliary_loss_clip": 0.01192837,
      "auxiliary_loss_mlp": 0.0102856,
      "balance_loss_clip": 1.0565691,
      "balance_loss_mlp": 1.02027488,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 1.8055562380016188,
      "language_loss": 0.75472301,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77693695,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.6319475173950195
    },
    {
      "auxiliary_loss_clip": 0.01197051,
      "auxiliary_loss_mlp": 0.01029697,
      "balance_loss_clip": 0.98244792,
      "balance_loss_mlp": 1.02082241,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 2.1349393552204265,
      "language_loss": 0.75203383,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77430129,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.674933433532715
    },
    {
      "auxiliary_loss_clip": 0.01192188,
      "auxiliary_loss_mlp": 0.01028578,
      "balance_loss_clip": 0.86459386,
      "balance_loss_mlp": 1.01978636,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 1.892801017126357,
      "language_loss": 0.77127671,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.79348439,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.795994520187378
    },
    {
      "auxiliary_loss_clip": 0.01175606,
      "auxiliary_loss_mlp": 0.01034833,
      "balance_loss_clip": 0.93803048,
      "balance_loss_mlp": 1.02653599,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 1.8305720080857248,
      "language_loss": 0.68329328,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.70539761,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 2.816646099090576
    },
    {
      "auxiliary_loss_clip": 0.01196315,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.06009221,
      "balance_loss_mlp": 1.02077723,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.6844116849903192,
      "language_loss": 0.69572794,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71798432,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 2.7026569843292236
    },
    {
      "auxiliary_loss_clip": 0.01198544,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 1.01944041,
      "balance_loss_mlp": 1.02012467,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 2.278845098824917,
      "language_loss": 0.72757155,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.7498489,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 2.6340079307556152
    },
    {
      "auxiliary_loss_clip": 0.01184331,
      "auxiliary_loss_mlp": 0.01030949,
      "balance_loss_clip": 0.94151866,
      "balance_loss_mlp": 1.02176428,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 2.207724220000312,
      "language_loss": 0.88209403,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90424687,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.754246234893799
    },
    {
      "auxiliary_loss_clip": 0.01185144,
      "auxiliary_loss_mlp": 0.01038706,
      "balance_loss_clip": 0.93931854,
      "balance_loss_mlp": 1.03002119,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.7399556403901721,
      "language_loss": 0.63066298,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65290141,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.6704344749450684
    },
    {
      "auxiliary_loss_clip": 0.01200875,
      "auxiliary_loss_mlp": 0.01032782,
      "balance_loss_clip": 1.06174505,
      "balance_loss_mlp": 1.02417517,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 3.3891216048678587,
      "language_loss": 0.71935737,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74169385,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 2.696444034576416
    },
    {
      "auxiliary_loss_clip": 0.01175855,
      "auxiliary_loss_mlp": 0.01031657,
      "balance_loss_clip": 0.93679368,
      "balance_loss_mlp": 1.02278161,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 2.3117594474003047,
      "language_loss": 0.74259222,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76466733,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.7571957111358643
    },
    {
      "auxiliary_loss_clip": 0.0117283,
      "auxiliary_loss_mlp": 0.0103763,
      "balance_loss_clip": 0.86048746,
      "balance_loss_mlp": 1.0289458,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 2.854338157690754,
      "language_loss": 0.6793232,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.70142782,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 2.8547556400299072
    },
    {
      "auxiliary_loss_clip": 0.0118277,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 0.94220161,
      "balance_loss_mlp": 1.02019775,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 1.7189962070515044,
      "language_loss": 0.69415009,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.71626747,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 4.885807752609253
    },
    {
      "auxiliary_loss_clip": 0.01201115,
      "auxiliary_loss_mlp": 0.01030152,
      "balance_loss_clip": 1.06350386,
      "balance_loss_mlp": 1.02174783,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 2.0994041779042814,
      "language_loss": 0.86873299,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.89104569,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 2.667719841003418
    },
    {
      "auxiliary_loss_clip": 0.01182917,
      "auxiliary_loss_mlp": 0.01125315,
      "balance_loss_clip": 0.82367504,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.5174678303972629,
      "language_loss": 0.76617205,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.78925437,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.883033514022827
    },
    {
      "auxiliary_loss_clip": 0.01196581,
      "auxiliary_loss_mlp": 0.01028649,
      "balance_loss_clip": 1.0198226,
      "balance_loss_mlp": 1.02005363,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 1.995838612143716,
      "language_loss": 0.65247703,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67472935,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.664654493331909
    },
    {
      "auxiliary_loss_clip": 0.01196058,
      "auxiliary_loss_mlp": 0.01034413,
      "balance_loss_clip": 1.0580337,
      "balance_loss_mlp": 1.02571702,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 2.066953564507281,
      "language_loss": 0.7974593,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.81976396,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 3.6511664390563965
    },
    {
      "auxiliary_loss_clip": 0.01191562,
      "auxiliary_loss_mlp": 0.01032709,
      "balance_loss_clip": 0.9825803,
      "balance_loss_mlp": 1.02450109,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 1.9264603038486494,
      "language_loss": 0.85677844,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.87902117,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.7857561111450195
    },
    {
      "auxiliary_loss_clip": 0.01196958,
      "auxiliary_loss_mlp": 0.0103306,
      "balance_loss_clip": 1.02207303,
      "balance_loss_mlp": 1.02481043,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 1.8312969596056625,
      "language_loss": 0.78813183,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.81043202,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 3.626903533935547
    },
    {
      "auxiliary_loss_clip": 0.01190788,
      "auxiliary_loss_mlp": 0.01125004,
      "balance_loss_clip": 0.94260252,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 1.8961802022539043,
      "language_loss": 0.82039142,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.84354937,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.777923583984375
    },
    {
      "auxiliary_loss_clip": 0.01186596,
      "auxiliary_loss_mlp": 0.01031841,
      "balance_loss_clip": 0.97698152,
      "balance_loss_mlp": 1.0233655,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 1.9462093402528338,
      "language_loss": 0.8008669,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.82305127,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.677694320678711
    },
    {
      "auxiliary_loss_clip": 0.01197443,
      "auxiliary_loss_mlp": 0.0102968,
      "balance_loss_clip": 1.02202964,
      "balance_loss_mlp": 1.02108479,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 1.9637077771420248,
      "language_loss": 0.71623278,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.73850393,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.748643159866333
    },
    {
      "auxiliary_loss_clip": 0.01197306,
      "auxiliary_loss_mlp": 0.01033793,
      "balance_loss_clip": 1.06016517,
      "balance_loss_mlp": 1.02534676,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 1.855211529343634,
      "language_loss": 0.77433467,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.79664564,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.671459913253784
    },
    {
      "auxiliary_loss_clip": 0.01196598,
      "auxiliary_loss_mlp": 0.01031352,
      "balance_loss_clip": 0.98269051,
      "balance_loss_mlp": 1.02322185,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 1.8202540649161492,
      "language_loss": 0.74310988,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76538938,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.7253682613372803
    },
    {
      "auxiliary_loss_clip": 0.01107639,
      "auxiliary_loss_mlp": 0.01001828,
      "balance_loss_clip": 1.00109529,
      "balance_loss_mlp": 0.99951535,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.3300454944373532,
      "language_loss": 0.64486903,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66596371,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.304776191711426
    },
    {
      "auxiliary_loss_clip": 0.01193089,
      "auxiliary_loss_mlp": 0.0102647,
      "balance_loss_clip": 0.93952578,
      "balance_loss_mlp": 1.01711178,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 2.6698339087976324,
      "language_loss": 0.70129335,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72348893,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.6683058738708496
    },
    {
      "auxiliary_loss_clip": 0.01104878,
      "auxiliary_loss_mlp": 0.01001365,
      "balance_loss_clip": 0.96047437,
      "balance_loss_mlp": 0.99906462,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.8886636860302936,
      "language_loss": 0.57780945,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.59887195,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.260206460952759
    },
    {
      "auxiliary_loss_clip": 0.01191713,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 0.90762037,
      "balance_loss_mlp": 1.02085137,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 2.131982579834035,
      "language_loss": 0.73965359,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.76186436,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.7752041816711426
    },
    {
      "auxiliary_loss_clip": 0.01187967,
      "auxiliary_loss_mlp": 0.0103472,
      "balance_loss_clip": 0.9066807,
      "balance_loss_mlp": 1.02674198,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 2.3086716943479666,
      "language_loss": 0.83717477,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.85940164,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.67897629737854
    },
    {
      "auxiliary_loss_clip": 0.01197888,
      "auxiliary_loss_mlp": 0.0102939,
      "balance_loss_clip": 1.02000105,
      "balance_loss_mlp": 1.02096152,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.0079428812635323,
      "language_loss": 0.71586025,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.73813307,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.6530604362487793
    },
    {
      "auxiliary_loss_clip": 0.01198877,
      "auxiliary_loss_mlp": 0.01032355,
      "balance_loss_clip": 0.98692679,
      "balance_loss_mlp": 1.02398705,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.055457729806609,
      "language_loss": 0.62144756,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.64375997,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 2.696984052658081
    },
    {
      "auxiliary_loss_clip": 0.01186364,
      "auxiliary_loss_mlp": 0.01027193,
      "balance_loss_clip": 0.97960472,
      "balance_loss_mlp": 1.01879478,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 1.9085113797136288,
      "language_loss": 0.82103515,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.84317064,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 2.6546437740325928
    },
    {
      "auxiliary_loss_clip": 0.01192044,
      "auxiliary_loss_mlp": 0.01028903,
      "balance_loss_clip": 1.02196014,
      "balance_loss_mlp": 1.02067745,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 1.8249982446349986,
      "language_loss": 0.63524079,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.65745032,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 2.68039608001709
    },
    {
      "auxiliary_loss_clip": 0.01183609,
      "auxiliary_loss_mlp": 0.01032579,
      "balance_loss_clip": 1.01927805,
      "balance_loss_mlp": 1.02465165,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.8530952076897342,
      "language_loss": 0.75185513,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77401698,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.7489030361175537
    },
    {
      "auxiliary_loss_clip": 0.0119069,
      "auxiliary_loss_mlp": 0.0103414,
      "balance_loss_clip": 0.9826948,
      "balance_loss_mlp": 1.02589655,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 2.3883585115496846,
      "language_loss": 0.80479801,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.82704628,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.6234164237976074
    },
    {
      "auxiliary_loss_clip": 0.01193383,
      "auxiliary_loss_mlp": 0.01028272,
      "balance_loss_clip": 1.02037621,
      "balance_loss_mlp": 1.02003431,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.6390006477508363,
      "language_loss": 0.69379467,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.71601129,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.6942172050476074
    },
    {
      "auxiliary_loss_clip": 0.01193304,
      "auxiliary_loss_mlp": 0.01024497,
      "balance_loss_clip": 0.97969097,
      "balance_loss_mlp": 1.01647985,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 1.8103296190123779,
      "language_loss": 0.64737153,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.66954958,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.6613540649414062
    },
    {
      "auxiliary_loss_clip": 0.01196405,
      "auxiliary_loss_mlp": 0.01029096,
      "balance_loss_clip": 1.05708718,
      "balance_loss_mlp": 1.02036095,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 2.1947113996421925,
      "language_loss": 0.81130469,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83355963,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 3.5104362964630127
    },
    {
      "auxiliary_loss_clip": 0.01183513,
      "auxiliary_loss_mlp": 0.01033934,
      "balance_loss_clip": 0.9421283,
      "balance_loss_mlp": 1.02592278,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 2.7021204773831613,
      "language_loss": 0.7907415,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81291598,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.721198320388794
    },
    {
      "auxiliary_loss_clip": 0.01195025,
      "auxiliary_loss_mlp": 0.0112458,
      "balance_loss_clip": 1.01949739,
      "balance_loss_mlp": 0.0,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 2.0822901051176377,
      "language_loss": 0.73085678,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.75405276,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 3.6494157314300537
    },
    {
      "auxiliary_loss_clip": 0.01196908,
      "auxiliary_loss_mlp": 0.01027352,
      "balance_loss_clip": 1.01832533,
      "balance_loss_mlp": 1.01971054,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 1.8110092808539255,
      "language_loss": 0.80725682,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.82949936,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 2.77630877494812
    },
    {
      "auxiliary_loss_clip": 0.01193984,
      "auxiliary_loss_mlp": 0.01028224,
      "balance_loss_clip": 1.02262187,
      "balance_loss_mlp": 1.02045774,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.6817880364481939,
      "language_loss": 0.78534162,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.80756372,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.693290948867798
    },
    {
      "auxiliary_loss_clip": 0.01183781,
      "auxiliary_loss_mlp": 0.01028375,
      "balance_loss_clip": 0.937392,
      "balance_loss_mlp": 1.02035785,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.5349687184577516,
      "language_loss": 0.72449291,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.74661446,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 3.6732029914855957
    },
    {
      "auxiliary_loss_clip": 0.01190386,
      "auxiliary_loss_mlp": 0.01029402,
      "balance_loss_clip": 1.02060938,
      "balance_loss_mlp": 1.02125418,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 1.7352783992254,
      "language_loss": 0.821684,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.84388185,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.7350072860717773
    },
    {
      "auxiliary_loss_clip": 0.01197179,
      "auxiliary_loss_mlp": 0.01123802,
      "balance_loss_clip": 1.01908612,
      "balance_loss_mlp": 0.0,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 2.710200936684309,
      "language_loss": 0.79479504,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.81800485,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.7243638038635254
    },
    {
      "auxiliary_loss_clip": 0.01109127,
      "auxiliary_loss_mlp": 0.01004854,
      "balance_loss_clip": 0.92907071,
      "balance_loss_mlp": 1.00241041,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 3.7418263837041312,
      "language_loss": 0.59218752,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.61332738,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 4.205641508102417
    },
    {
      "auxiliary_loss_clip": 0.0118812,
      "auxiliary_loss_mlp": 0.01032696,
      "balance_loss_clip": 0.9788928,
      "balance_loss_mlp": 1.02370167,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.5908170462059326,
      "language_loss": 0.70298815,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.72519636,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.793691873550415
    },
    {
      "auxiliary_loss_clip": 0.01182189,
      "auxiliary_loss_mlp": 0.01033278,
      "balance_loss_clip": 0.93849343,
      "balance_loss_mlp": 1.02546394,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 1.6799863568827356,
      "language_loss": 0.82829511,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.8504498,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.7942137718200684
    },
    {
      "auxiliary_loss_clip": 0.011951,
      "auxiliary_loss_mlp": 0.01030801,
      "balance_loss_clip": 1.05827999,
      "balance_loss_mlp": 1.02229571,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 1.6657592642816457,
      "language_loss": 0.73036027,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75261927,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.6844985485076904
    },
    {
      "auxiliary_loss_clip": 0.01183282,
      "auxiliary_loss_mlp": 0.01027854,
      "balance_loss_clip": 0.93968433,
      "balance_loss_mlp": 1.01992035,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 1.7460865328048656,
      "language_loss": 0.76758444,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.78969574,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.8534352779388428
    },
    {
      "auxiliary_loss_clip": 0.01193128,
      "auxiliary_loss_mlp": 0.01030292,
      "balance_loss_clip": 1.02189755,
      "balance_loss_mlp": 1.022174,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 2.021357223318659,
      "language_loss": 0.79315758,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.81539178,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.663804769515991
    },
    {
      "auxiliary_loss_clip": 0.01191507,
      "auxiliary_loss_mlp": 0.01022202,
      "balance_loss_clip": 1.0191046,
      "balance_loss_mlp": 1.01484704,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 1.8393627795540473,
      "language_loss": 0.80766886,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.82980597,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.638836622238159
    },
    {
      "auxiliary_loss_clip": 0.01195793,
      "auxiliary_loss_mlp": 0.01026498,
      "balance_loss_clip": 1.02080834,
      "balance_loss_mlp": 1.01855898,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 2.289988571574782,
      "language_loss": 0.77194446,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.7941674,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.6504392623901367
    },
    {
      "auxiliary_loss_clip": 0.01191106,
      "auxiliary_loss_mlp": 0.01030307,
      "balance_loss_clip": 1.01923203,
      "balance_loss_mlp": 1.02211106,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 5.23599348514322,
      "language_loss": 0.83220208,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85441619,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.735863208770752
    },
    {
      "auxiliary_loss_clip": 0.01189523,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.02052987,
      "balance_loss_mlp": 1.01672459,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 1.4767136393933273,
      "language_loss": 0.88528252,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90742534,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.7690858840942383
    },
    {
      "auxiliary_loss_clip": 0.01190844,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 0.98124111,
      "balance_loss_mlp": 1.01766789,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.1226095200347093,
      "language_loss": 0.88808686,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.91025138,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.83487868309021
    },
    {
      "auxiliary_loss_clip": 0.01183136,
      "auxiliary_loss_mlp": 0.01032085,
      "balance_loss_clip": 0.97711229,
      "balance_loss_mlp": 1.02354944,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 2.614832318081303,
      "language_loss": 0.81304479,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.83519697,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 2.698054552078247
    },
    {
      "auxiliary_loss_clip": 0.01192178,
      "auxiliary_loss_mlp": 0.01029161,
      "balance_loss_clip": 1.01891506,
      "balance_loss_mlp": 1.02117383,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 2.35938605351568,
      "language_loss": 0.71518612,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73739952,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 2.863762140274048
    },
    {
      "auxiliary_loss_clip": 0.01194338,
      "auxiliary_loss_mlp": 0.01031931,
      "balance_loss_clip": 0.94209254,
      "balance_loss_mlp": 1.02349114,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 2.102132243826137,
      "language_loss": 0.89659864,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.91886139,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 2.688711643218994
    },
    {
      "auxiliary_loss_clip": 0.01101038,
      "auxiliary_loss_mlp": 0.01006861,
      "balance_loss_clip": 0.87995303,
      "balance_loss_mlp": 1.00427425,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.6930086758900473,
      "language_loss": 0.56652182,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58760089,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.5504586696624756
    },
    {
      "auxiliary_loss_clip": 0.0119325,
      "auxiliary_loss_mlp": 0.01029215,
      "balance_loss_clip": 0.90268564,
      "balance_loss_mlp": 1.02151382,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 2.2226478044274627,
      "language_loss": 0.66988742,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69211209,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 2.807157516479492
    },
    {
      "auxiliary_loss_clip": 0.01181341,
      "auxiliary_loss_mlp": 0.0102953,
      "balance_loss_clip": 0.94120336,
      "balance_loss_mlp": 1.02173924,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 25.099257230307206,
      "language_loss": 0.76060176,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.78271049,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.729779005050659
    },
    {
      "auxiliary_loss_clip": 0.01194512,
      "auxiliary_loss_mlp": 0.01124376,
      "balance_loss_clip": 1.05585194,
      "balance_loss_mlp": 0.0,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.081740723899189,
      "language_loss": 0.82152665,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.8447156,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.650230884552002
    },
    {
      "auxiliary_loss_clip": 0.01178759,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.01416147,
      "balance_loss_mlp": 1.02010667,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 2.0277963456637735,
      "language_loss": 0.83046079,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85253179,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 3.627150535583496
    },
    {
      "auxiliary_loss_clip": 0.0119263,
      "auxiliary_loss_mlp": 0.01029411,
      "balance_loss_clip": 0.94250655,
      "balance_loss_mlp": 1.02049363,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 1.5546617388964474,
      "language_loss": 0.79030681,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81252718,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 2.75616455078125
    },
    {
      "auxiliary_loss_clip": 0.01193688,
      "auxiliary_loss_mlp": 0.01025034,
      "balance_loss_clip": 1.01885307,
      "balance_loss_mlp": 1.01603377,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 4.219168050334588,
      "language_loss": 0.77671981,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79890704,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 3.708709716796875
    },
    {
      "auxiliary_loss_clip": 0.01191882,
      "auxiliary_loss_mlp": 0.01124273,
      "balance_loss_clip": 1.05645347,
      "balance_loss_mlp": 0.0,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 1.838265565256358,
      "language_loss": 0.73963284,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.76279438,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.5988757610321045
    },
    {
      "auxiliary_loss_clip": 0.01189635,
      "auxiliary_loss_mlp": 0.01028463,
      "balance_loss_clip": 0.97799885,
      "balance_loss_mlp": 1.02038383,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 1.8529368352347277,
      "language_loss": 0.7895422,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.81172311,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.7212188243865967
    },
    {
      "auxiliary_loss_clip": 0.01184165,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 0.94099796,
      "balance_loss_mlp": 1.01658297,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 2.4305182457952372,
      "language_loss": 0.68427372,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.70636022,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.6876907348632812
    },
    {
      "auxiliary_loss_clip": 0.01183292,
      "auxiliary_loss_mlp": 0.01031713,
      "balance_loss_clip": 0.97767985,
      "balance_loss_mlp": 1.02268267,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 7.860459760903073,
      "language_loss": 0.79046559,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.81261563,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 3.65718412399292
    },
    {
      "auxiliary_loss_clip": 0.01188833,
      "auxiliary_loss_mlp": 0.01029313,
      "balance_loss_clip": 0.98003799,
      "balance_loss_mlp": 1.02203536,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 1.9808642026346157,
      "language_loss": 0.78810954,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.81029099,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.693861484527588
    },
    {
      "auxiliary_loss_clip": 0.01183814,
      "auxiliary_loss_mlp": 0.01034314,
      "balance_loss_clip": 0.978688,
      "balance_loss_mlp": 1.02617252,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 1.791846488065643,
      "language_loss": 0.76448226,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.78666353,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 3.6374473571777344
    },
    {
      "auxiliary_loss_clip": 0.01188339,
      "auxiliary_loss_mlp": 0.01024229,
      "balance_loss_clip": 1.01419199,
      "balance_loss_mlp": 1.01633787,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 2.3553095931192147,
      "language_loss": 0.76330835,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.78543401,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.702526330947876
    },
    {
      "auxiliary_loss_clip": 0.01178935,
      "auxiliary_loss_mlp": 0.0103205,
      "balance_loss_clip": 0.97640383,
      "balance_loss_mlp": 1.02349043,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 3.256448430226864,
      "language_loss": 0.74202925,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76413912,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.713094711303711
    },
    {
      "auxiliary_loss_clip": 0.01095616,
      "auxiliary_loss_mlp": 0.01001854,
      "balance_loss_clip": 1.02770913,
      "balance_loss_mlp": 0.99939823,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7743467821642804,
      "language_loss": 0.57772487,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59869957,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 3.0405852794647217
    },
    {
      "auxiliary_loss_clip": 0.01189738,
      "auxiliary_loss_mlp": 0.01025058,
      "balance_loss_clip": 0.90122432,
      "balance_loss_mlp": 1.01707399,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 1.5131106644951753,
      "language_loss": 0.78164721,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.80379516,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.9558331966400146
    },
    {
      "auxiliary_loss_clip": 0.01190201,
      "auxiliary_loss_mlp": 0.01035199,
      "balance_loss_clip": 0.98254979,
      "balance_loss_mlp": 1.0269022,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 7.788929911000162,
      "language_loss": 0.57466304,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.59691703,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.9228222370147705
    },
    {
      "auxiliary_loss_clip": 0.01195487,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 0.98044968,
      "balance_loss_mlp": 1.02360392,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 1.8547205394389599,
      "language_loss": 0.69527566,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.71755153,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.6913869380950928
    },
    {
      "auxiliary_loss_clip": 0.01190693,
      "auxiliary_loss_mlp": 0.01027958,
      "balance_loss_clip": 0.97611386,
      "balance_loss_mlp": 1.02014422,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.052696033791829,
      "language_loss": 0.77696824,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79915476,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.629202127456665
    },
    {
      "auxiliary_loss_clip": 0.01183688,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 0.97722059,
      "balance_loss_mlp": 1.02081096,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 1.709474040225576,
      "language_loss": 0.91485667,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.93698108,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.67492413520813
    },
    {
      "auxiliary_loss_clip": 0.01190166,
      "auxiliary_loss_mlp": 0.0103192,
      "balance_loss_clip": 0.97993922,
      "balance_loss_mlp": 1.02423143,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 1.7227008191750428,
      "language_loss": 0.74980533,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77202618,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.6855404376983643
    },
    {
      "auxiliary_loss_clip": 0.01192353,
      "auxiliary_loss_mlp": 0.01025858,
      "balance_loss_clip": 0.90339935,
      "balance_loss_mlp": 1.01768601,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.136602216657118,
      "language_loss": 0.75453943,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.77672148,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.756228446960449
    },
    {
      "auxiliary_loss_clip": 0.01186367,
      "auxiliary_loss_mlp": 0.01031344,
      "balance_loss_clip": 0.9431054,
      "balance_loss_mlp": 1.02286196,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 1.7840595787585867,
      "language_loss": 0.85161281,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.87378991,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 2.760556697845459
    },
    {
      "auxiliary_loss_clip": 0.0119533,
      "auxiliary_loss_mlp": 0.01033648,
      "balance_loss_clip": 1.05699015,
      "balance_loss_mlp": 1.02476728,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 2.237801340552917,
      "language_loss": 0.73831475,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.76060456,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 2.6476681232452393
    },
    {
      "auxiliary_loss_clip": 0.01196373,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.02001667,
      "balance_loss_mlp": 1.01536226,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 2.4469893492524113,
      "language_loss": 0.72265035,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74485177,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.6212058067321777
    },
    {
      "auxiliary_loss_clip": 0.01187687,
      "auxiliary_loss_mlp": 0.01029529,
      "balance_loss_clip": 0.97880483,
      "balance_loss_mlp": 1.02185178,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 3.701941530428575,
      "language_loss": 0.73270524,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.75487733,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.8296058177948
    },
    {
      "auxiliary_loss_clip": 0.01178729,
      "auxiliary_loss_mlp": 0.01024084,
      "balance_loss_clip": 0.90221101,
      "balance_loss_mlp": 1.01665092,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 1.697771285944049,
      "language_loss": 0.72910082,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.75112897,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 2.754863739013672
    },
    {
      "auxiliary_loss_clip": 0.01172585,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 0.85938954,
      "balance_loss_mlp": 1.01810563,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 1.6753703741753938,
      "language_loss": 0.8174603,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.83944643,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.8170034885406494
    },
    {
      "auxiliary_loss_clip": 0.01195715,
      "auxiliary_loss_mlp": 0.01034476,
      "balance_loss_clip": 1.01863909,
      "balance_loss_mlp": 1.02636981,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 1.923162360048908,
      "language_loss": 0.72679085,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.74909276,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 3.6675608158111572
    },
    {
      "auxiliary_loss_clip": 0.01187672,
      "auxiliary_loss_mlp": 0.0102858,
      "balance_loss_clip": 0.97988796,
      "balance_loss_mlp": 1.0206883,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 1.899726743310442,
      "language_loss": 0.84637487,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.86853743,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 2.7085087299346924
    },
    {
      "auxiliary_loss_clip": 0.01193979,
      "auxiliary_loss_mlp": 0.01029508,
      "balance_loss_clip": 1.05572784,
      "balance_loss_mlp": 1.02130079,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.343876697319984,
      "language_loss": 0.7679528,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.79018772,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 2.6238784790039062
    },
    {
      "auxiliary_loss_clip": 0.01188889,
      "auxiliary_loss_mlp": 0.01124137,
      "balance_loss_clip": 1.01874363,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.010091874229172,
      "language_loss": 0.82634485,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84947509,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 3.769545316696167
    },
    {
      "auxiliary_loss_clip": 0.01179815,
      "auxiliary_loss_mlp": 0.01026722,
      "balance_loss_clip": 0.9369117,
      "balance_loss_mlp": 1.0186547,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 1.9136842995045333,
      "language_loss": 0.73186839,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75393379,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.727431058883667
    },
    {
      "auxiliary_loss_clip": 0.01190827,
      "auxiliary_loss_mlp": 0.01031059,
      "balance_loss_clip": 1.01834321,
      "balance_loss_mlp": 1.02264261,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 1.7296369699521799,
      "language_loss": 0.81404638,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83626521,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.7008419036865234
    },
    {
      "auxiliary_loss_clip": 0.01178838,
      "auxiliary_loss_mlp": 0.01124075,
      "balance_loss_clip": 0.90210688,
      "balance_loss_mlp": 0.0,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 2.070760231022463,
      "language_loss": 0.8738873,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89691639,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.860365152359009
    },
    {
      "auxiliary_loss_clip": 0.01164705,
      "auxiliary_loss_mlp": 0.01034056,
      "balance_loss_clip": 0.85953271,
      "balance_loss_mlp": 1.02611935,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.4793249070920016,
      "language_loss": 0.88209385,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.90408146,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 3.8138105869293213
    },
    {
      "auxiliary_loss_clip": 0.01191159,
      "auxiliary_loss_mlp": 0.01035356,
      "balance_loss_clip": 0.93886697,
      "balance_loss_mlp": 1.02633774,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 1.96462051088435,
      "language_loss": 0.80083752,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82310265,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.728410005569458
    },
    {
      "auxiliary_loss_clip": 0.01191576,
      "auxiliary_loss_mlp": 0.0102968,
      "balance_loss_clip": 1.02121103,
      "balance_loss_mlp": 1.02194905,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 1.850305182533223,
      "language_loss": 0.85689044,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.879103,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 3.50809645652771
    },
    {
      "auxiliary_loss_clip": 0.0119096,
      "auxiliary_loss_mlp": 0.01030297,
      "balance_loss_clip": 0.93784374,
      "balance_loss_mlp": 1.02214956,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 1.670192548916342,
      "language_loss": 0.9034276,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.92564017,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.734013557434082
    },
    {
      "auxiliary_loss_clip": 0.01184082,
      "auxiliary_loss_mlp": 0.01032533,
      "balance_loss_clip": 0.97671795,
      "balance_loss_mlp": 1.02474225,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 1.9589017271876947,
      "language_loss": 0.6971038,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.71926999,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 2.643296480178833
    },
    {
      "auxiliary_loss_clip": 0.01183088,
      "auxiliary_loss_mlp": 0.0102571,
      "balance_loss_clip": 0.89810765,
      "balance_loss_mlp": 1.01772261,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 1.952245294585522,
      "language_loss": 0.79727489,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.81936282,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.8337018489837646
    },
    {
      "auxiliary_loss_clip": 0.01187305,
      "auxiliary_loss_mlp": 0.01033649,
      "balance_loss_clip": 1.02025247,
      "balance_loss_mlp": 1.02557278,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 1.448764464263474,
      "language_loss": 0.67541671,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.69762623,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.845961570739746
    },
    {
      "auxiliary_loss_clip": 0.01189605,
      "auxiliary_loss_mlp": 0.01032822,
      "balance_loss_clip": 0.97646332,
      "balance_loss_mlp": 1.02477503,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 1.9383278294314412,
      "language_loss": 0.78724611,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.80947036,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.735365629196167
    },
    {
      "auxiliary_loss_clip": 0.01185696,
      "auxiliary_loss_mlp": 0.01029082,
      "balance_loss_clip": 0.94484079,
      "balance_loss_mlp": 1.02112484,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 1.7764204972378257,
      "language_loss": 0.79711258,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.81926042,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.7430901527404785
    },
    {
      "auxiliary_loss_clip": 0.0119802,
      "auxiliary_loss_mlp": 0.01033301,
      "balance_loss_clip": 1.0594877,
      "balance_loss_mlp": 1.02433658,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 1.888958604607165,
      "language_loss": 0.7506932,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77300644,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.635979175567627
    },
    {
      "auxiliary_loss_clip": 0.01187379,
      "auxiliary_loss_mlp": 0.01032593,
      "balance_loss_clip": 0.90578496,
      "balance_loss_mlp": 1.02417731,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.5987242994943682,
      "language_loss": 0.76654041,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78874016,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.8504843711853027
    },
    {
      "auxiliary_loss_clip": 0.01189145,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 0.93981671,
      "balance_loss_mlp": 1.02138638,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 1.8857098652301454,
      "language_loss": 0.71764082,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73983043,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.7805371284484863
    },
    {
      "auxiliary_loss_clip": 0.01195851,
      "auxiliary_loss_mlp": 0.01034177,
      "balance_loss_clip": 1.05992544,
      "balance_loss_mlp": 1.02607703,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 1.645402040741893,
      "language_loss": 0.80905378,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83135414,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 2.6881728172302246
    },
    {
      "auxiliary_loss_clip": 0.01188388,
      "auxiliary_loss_mlp": 0.01033094,
      "balance_loss_clip": 1.01792717,
      "balance_loss_mlp": 1.02486849,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 1.70997828162107,
      "language_loss": 0.75902271,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.78123748,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 2.7326736450195312
    },
    {
      "auxiliary_loss_clip": 0.01178552,
      "auxiliary_loss_mlp": 0.0112482,
      "balance_loss_clip": 0.86527658,
      "balance_loss_mlp": 0.0,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 1.9650554342860795,
      "language_loss": 0.80167079,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.82470453,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 2.829901695251465
    },
    {
      "auxiliary_loss_clip": 0.01176503,
      "auxiliary_loss_mlp": 0.01027093,
      "balance_loss_clip": 0.93575048,
      "balance_loss_mlp": 1.01918936,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.758595191277045,
      "language_loss": 0.8291865,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85122246,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.709298849105835
    },
    {
      "auxiliary_loss_clip": 0.0118814,
      "auxiliary_loss_mlp": 0.01024078,
      "balance_loss_clip": 0.97710276,
      "balance_loss_mlp": 1.01646352,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.5067953787719188,
      "language_loss": 0.825409,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84753114,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.721916913986206
    },
    {
      "auxiliary_loss_clip": 0.01189799,
      "auxiliary_loss_mlp": 0.0103904,
      "balance_loss_clip": 0.98272252,
      "balance_loss_mlp": 1.03071964,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.552516989691585,
      "language_loss": 0.7105732,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.73286164,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 2.734168767929077
    },
    {
      "auxiliary_loss_clip": 0.01194134,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 0.94037277,
      "balance_loss_mlp": 1.01704276,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 1.7385025677942143,
      "language_loss": 0.82542276,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.84761453,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 2.783940076828003
    },
    {
      "auxiliary_loss_clip": 0.01197226,
      "auxiliary_loss_mlp": 0.01027576,
      "balance_loss_clip": 0.9793824,
      "balance_loss_mlp": 1.01897478,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.3631159422666785,
      "language_loss": 0.76004934,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.78229737,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 3.6715846061706543
    },
    {
      "auxiliary_loss_clip": 0.01194478,
      "auxiliary_loss_mlp": 0.01033023,
      "balance_loss_clip": 1.05662966,
      "balance_loss_mlp": 1.02519071,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.7050211388742749,
      "language_loss": 0.84337187,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86564696,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 2.679966926574707
    },
    {
      "auxiliary_loss_clip": 0.01193023,
      "auxiliary_loss_mlp": 0.01023932,
      "balance_loss_clip": 1.01935458,
      "balance_loss_mlp": 1.01591563,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 1.849800002627672,
      "language_loss": 0.71421874,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.73638833,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 2.710320472717285
    },
    {
      "auxiliary_loss_clip": 0.01197473,
      "auxiliary_loss_mlp": 0.01028877,
      "balance_loss_clip": 0.98021436,
      "balance_loss_mlp": 1.02004361,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 1.9094314962349255,
      "language_loss": 0.76568663,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.7879501,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 3.7176456451416016
    },
    {
      "auxiliary_loss_clip": 0.0117468,
      "auxiliary_loss_mlp": 0.01124812,
      "balance_loss_clip": 0.93704224,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 2.0152913427712202,
      "language_loss": 0.77853024,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80152524,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 2.7467124462127686
    },
    {
      "auxiliary_loss_clip": 0.01193187,
      "auxiliary_loss_mlp": 0.01031598,
      "balance_loss_clip": 0.98143804,
      "balance_loss_mlp": 1.02261579,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 1.6620633789930717,
      "language_loss": 0.69882727,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72107518,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.871314525604248
    },
    {
      "auxiliary_loss_clip": 0.01185749,
      "auxiliary_loss_mlp": 0.01028476,
      "balance_loss_clip": 0.9022187,
      "balance_loss_mlp": 1.02012503,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 2.577882912474391,
      "language_loss": 0.64141041,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.66355264,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.7615203857421875
    },
    {
      "auxiliary_loss_clip": 0.01184468,
      "auxiliary_loss_mlp": 0.01031495,
      "balance_loss_clip": 0.90034896,
      "balance_loss_mlp": 1.02315593,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 2.6113059234226728,
      "language_loss": 0.76131588,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78347552,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 3.6787219047546387
    },
    {
      "auxiliary_loss_clip": 0.0118286,
      "auxiliary_loss_mlp": 0.01025459,
      "balance_loss_clip": 0.97684163,
      "balance_loss_mlp": 1.0173285,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 2.0243707910478292,
      "language_loss": 0.66393113,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68601435,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 3.6329102516174316
    },
    {
      "auxiliary_loss_clip": 0.01195558,
      "auxiliary_loss_mlp": 0.01026483,
      "balance_loss_clip": 1.05904555,
      "balance_loss_mlp": 1.01862693,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 1.6083889778760299,
      "language_loss": 0.81623644,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.83845681,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.6362075805664062
    },
    {
      "auxiliary_loss_clip": 0.01199411,
      "auxiliary_loss_mlp": 0.01036368,
      "balance_loss_clip": 0.98118937,
      "balance_loss_mlp": 1.02796412,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 1.8587926015549934,
      "language_loss": 0.80039328,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82275105,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.743699073791504
    },
    {
      "auxiliary_loss_clip": 0.01192585,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.01994681,
      "balance_loss_mlp": 1.02073336,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 1.7019225489927348,
      "language_loss": 0.65323436,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67545211,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.7106809616088867
    },
    {
      "auxiliary_loss_clip": 0.01189158,
      "auxiliary_loss_mlp": 0.01030039,
      "balance_loss_clip": 1.05389857,
      "balance_loss_mlp": 1.02199817,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 1.7938440885839082,
      "language_loss": 0.85372066,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87591267,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.6580538749694824
    },
    {
      "auxiliary_loss_clip": 0.0118241,
      "auxiliary_loss_mlp": 0.01124137,
      "balance_loss_clip": 0.90251315,
      "balance_loss_mlp": 0.0,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 1.8761100851303172,
      "language_loss": 0.78042543,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.80349094,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.768232822418213
    },
    {
      "auxiliary_loss_clip": 0.01185247,
      "auxiliary_loss_mlp": 0.01029073,
      "balance_loss_clip": 0.93877035,
      "balance_loss_mlp": 1.02168226,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 1.7232678525412723,
      "language_loss": 0.78679931,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.8089425,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.8014180660247803
    },
    {
      "auxiliary_loss_clip": 0.01195499,
      "auxiliary_loss_mlp": 0.01030959,
      "balance_loss_clip": 1.05848455,
      "balance_loss_mlp": 1.02301967,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 2.649058679694957,
      "language_loss": 0.74347788,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.76574248,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.6964752674102783
    },
    {
      "auxiliary_loss_clip": 0.01183426,
      "auxiliary_loss_mlp": 0.01026892,
      "balance_loss_clip": 0.90521836,
      "balance_loss_mlp": 1.01870275,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 1.665319673448977,
      "language_loss": 0.68611974,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.70822293,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.81390118598938
    },
    {
      "auxiliary_loss_clip": 0.01193753,
      "auxiliary_loss_mlp": 0.01032257,
      "balance_loss_clip": 0.98201984,
      "balance_loss_mlp": 1.02417469,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 1.6122119489155116,
      "language_loss": 0.82807118,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.85033131,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.787149429321289
    },
    {
      "auxiliary_loss_clip": 0.011942,
      "auxiliary_loss_mlp": 0.01040181,
      "balance_loss_clip": 1.05611968,
      "balance_loss_mlp": 1.03141332,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 2.0859550370769266,
      "language_loss": 0.7377646,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.76010847,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.550755023956299
    },
    {
      "auxiliary_loss_clip": 0.01194151,
      "auxiliary_loss_mlp": 0.01028475,
      "balance_loss_clip": 1.01926851,
      "balance_loss_mlp": 1.02023137,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 1.6861129873707221,
      "language_loss": 0.6826973,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.70492363,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.716369390487671
    },
    {
      "auxiliary_loss_clip": 0.0119113,
      "auxiliary_loss_mlp": 0.010343,
      "balance_loss_clip": 0.98146921,
      "balance_loss_mlp": 1.02708781,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.578705819173538,
      "language_loss": 0.84134412,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86359841,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 2.7357680797576904
    },
    {
      "auxiliary_loss_clip": 0.01192027,
      "auxiliary_loss_mlp": 0.01032455,
      "balance_loss_clip": 1.0184983,
      "balance_loss_mlp": 1.02406287,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.1824457119791143,
      "language_loss": 0.82361567,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.84586048,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.6025969982147217
    },
    {
      "auxiliary_loss_clip": 0.01186381,
      "auxiliary_loss_mlp": 0.01125049,
      "balance_loss_clip": 0.82514727,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 4.849006586109743,
      "language_loss": 0.82557213,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.8486864,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 2.877042770385742
    },
    {
      "auxiliary_loss_clip": 0.01172235,
      "auxiliary_loss_mlp": 0.01030187,
      "balance_loss_clip": 0.97409564,
      "balance_loss_mlp": 1.02179503,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 1.935420458156833,
      "language_loss": 0.80152279,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.82354707,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 2.8850274085998535
    },
    {
      "auxiliary_loss_clip": 0.01189729,
      "auxiliary_loss_mlp": 0.01124286,
      "balance_loss_clip": 1.05581141,
      "balance_loss_mlp": 0.0,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 1.7618090365000294,
      "language_loss": 0.76958722,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.79272735,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.9655380249023438
    },
    {
      "auxiliary_loss_clip": 0.0109788,
      "auxiliary_loss_mlp": 0.010058,
      "balance_loss_clip": 1.02786827,
      "balance_loss_mlp": 1.00336838,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.7560404238355903,
      "language_loss": 0.60075045,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62178725,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 3.2212774753570557
    },
    {
      "auxiliary_loss_clip": 0.01196638,
      "auxiliary_loss_mlp": 0.01034609,
      "balance_loss_clip": 0.94462848,
      "balance_loss_mlp": 1.02652061,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 1.8371523334849433,
      "language_loss": 0.63787502,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66018748,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 3.8090691566467285
    },
    {
      "auxiliary_loss_clip": 0.01191825,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.05490601,
      "balance_loss_mlp": 1.01820374,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 2.077620140604334,
      "language_loss": 0.69639397,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.71857631,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.650858163833618
    },
    {
      "auxiliary_loss_clip": 0.01189319,
      "auxiliary_loss_mlp": 0.010348,
      "balance_loss_clip": 0.97950029,
      "balance_loss_mlp": 1.02673578,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 1.9398886906044952,
      "language_loss": 0.70671427,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72895545,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 2.757361888885498
    },
    {
      "auxiliary_loss_clip": 0.01181927,
      "auxiliary_loss_mlp": 0.01029397,
      "balance_loss_clip": 0.94181907,
      "balance_loss_mlp": 1.02082014,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.3672556654940418,
      "language_loss": 0.86693549,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.8890487,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 3.6829121112823486
    },
    {
      "auxiliary_loss_clip": 0.01186897,
      "auxiliary_loss_mlp": 0.01032537,
      "balance_loss_clip": 1.01708531,
      "balance_loss_mlp": 1.024544,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 2.0560215393290293,
      "language_loss": 0.77492481,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.79711914,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.650960922241211
    },
    {
      "auxiliary_loss_clip": 0.01193754,
      "auxiliary_loss_mlp": 0.01033757,
      "balance_loss_clip": 1.01784778,
      "balance_loss_mlp": 1.02579403,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 1.8342483216749559,
      "language_loss": 0.71741962,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.73969471,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.6538379192352295
    },
    {
      "auxiliary_loss_clip": 0.01096375,
      "auxiliary_loss_mlp": 0.011197,
      "balance_loss_clip": 0.99052,
      "balance_loss_mlp": 0.0,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 1.0448026367546828,
      "language_loss": 0.64240777,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.66456854,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.340041160583496
    },
    {
      "auxiliary_loss_clip": 0.01168787,
      "auxiliary_loss_mlp": 0.01030129,
      "balance_loss_clip": 0.86080915,
      "balance_loss_mlp": 1.02164161,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.452229004537399,
      "language_loss": 0.78299546,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.80498457,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.945021629333496
    },
    {
      "auxiliary_loss_clip": 0.01180194,
      "auxiliary_loss_mlp": 0.01025284,
      "balance_loss_clip": 0.86044312,
      "balance_loss_mlp": 1.01771402,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 1.6109705153635812,
      "language_loss": 0.82643461,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.8484894,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 4.71534276008606
    },
    {
      "auxiliary_loss_clip": 0.01190055,
      "auxiliary_loss_mlp": 0.01033534,
      "balance_loss_clip": 0.97845137,
      "balance_loss_mlp": 1.02544606,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.6390779081320288,
      "language_loss": 0.76042479,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.7826606,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.745900869369507
    },
    {
      "auxiliary_loss_clip": 0.01199653,
      "auxiliary_loss_mlp": 0.01041145,
      "balance_loss_clip": 1.02024531,
      "balance_loss_mlp": 1.03225183,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 2.0309655431945584,
      "language_loss": 0.76527274,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.78768075,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.633462429046631
    },
    {
      "auxiliary_loss_clip": 0.01180181,
      "auxiliary_loss_mlp": 0.01031202,
      "balance_loss_clip": 0.9071753,
      "balance_loss_mlp": 1.02372801,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 2.0579116577904446,
      "language_loss": 0.92465842,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94677234,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.751129627227783
    },
    {
      "auxiliary_loss_clip": 0.01181337,
      "auxiliary_loss_mlp": 0.01027376,
      "balance_loss_clip": 0.90199167,
      "balance_loss_mlp": 1.01992536,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 1.8654249836776824,
      "language_loss": 0.77030361,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.7923907,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.737323522567749
    },
    {
      "auxiliary_loss_clip": 0.01180907,
      "auxiliary_loss_mlp": 0.01034385,
      "balance_loss_clip": 1.01429987,
      "balance_loss_mlp": 1.02624846,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 1.74359312889328,
      "language_loss": 0.7635752,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.7857281,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.7106337547302246
    },
    {
      "auxiliary_loss_clip": 0.01183916,
      "auxiliary_loss_mlp": 0.01029791,
      "balance_loss_clip": 0.98166907,
      "balance_loss_mlp": 1.02153563,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 2.226696804875745,
      "language_loss": 0.70472145,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72685856,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.733063220977783
    },
    {
      "auxiliary_loss_clip": 0.01185907,
      "auxiliary_loss_mlp": 0.01029439,
      "balance_loss_clip": 0.86119223,
      "balance_loss_mlp": 1.02064705,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 1.9735883433087484,
      "language_loss": 0.78979647,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.81194991,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.8347015380859375
    },
    {
      "auxiliary_loss_clip": 0.01193385,
      "auxiliary_loss_mlp": 0.01024058,
      "balance_loss_clip": 0.97585821,
      "balance_loss_mlp": 1.01631522,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 1.694639553824735,
      "language_loss": 0.79399848,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.8161729,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.7115917205810547
    },
    {
      "auxiliary_loss_clip": 0.01112778,
      "auxiliary_loss_mlp": 0.01000484,
      "balance_loss_clip": 0.88292015,
      "balance_loss_mlp": 0.99783748,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7169879062152581,
      "language_loss": 0.56537819,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.58651078,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.306030750274658
    },
    {
      "auxiliary_loss_clip": 0.01177841,
      "auxiliary_loss_mlp": 0.0102959,
      "balance_loss_clip": 0.94059914,
      "balance_loss_mlp": 1.02187681,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 5.58630347167216,
      "language_loss": 0.77387512,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.79594946,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 2.797353982925415
    },
    {
      "auxiliary_loss_clip": 0.01192659,
      "auxiliary_loss_mlp": 0.01029801,
      "balance_loss_clip": 0.97678626,
      "balance_loss_mlp": 1.02093172,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 2.065921754587545,
      "language_loss": 0.74951494,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.77173954,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.706639528274536
    },
    {
      "auxiliary_loss_clip": 0.01101889,
      "auxiliary_loss_mlp": 0.010023,
      "balance_loss_clip": 0.9545511,
      "balance_loss_mlp": 0.99974841,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.8592957139551254,
      "language_loss": 0.60867155,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.62971342,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 3.2901391983032227
    },
    {
      "auxiliary_loss_clip": 0.01192508,
      "auxiliary_loss_mlp": 0.0103111,
      "balance_loss_clip": 1.0561161,
      "balance_loss_mlp": 1.0230037,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 1.7176013596611521,
      "language_loss": 0.66771722,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.68995339,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.5982024669647217
    },
    {
      "auxiliary_loss_clip": 0.01189181,
      "auxiliary_loss_mlp": 0.01024116,
      "balance_loss_clip": 0.93678594,
      "balance_loss_mlp": 1.01656413,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 2.0535514488872666,
      "language_loss": 0.79511297,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.81724596,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 2.7764198780059814
    },
    {
      "auxiliary_loss_clip": 0.01194825,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.01926017,
      "balance_loss_mlp": 1.02444744,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 1.9798521687825275,
      "language_loss": 0.86384416,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.88611692,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.727426052093506
    },
    {
      "auxiliary_loss_clip": 0.01188901,
      "auxiliary_loss_mlp": 0.01030184,
      "balance_loss_clip": 0.97983277,
      "balance_loss_mlp": 1.02182162,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 2.4576037905450088,
      "language_loss": 0.81213558,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.83432639,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.6871182918548584
    },
    {
      "auxiliary_loss_clip": 0.01187521,
      "auxiliary_loss_mlp": 0.01030772,
      "balance_loss_clip": 0.90038848,
      "balance_loss_mlp": 1.0231725,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 2.166241864791019,
      "language_loss": 0.75459456,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.77677751,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 2.761700391769409
    },
    {
      "auxiliary_loss_clip": 0.01097697,
      "auxiliary_loss_mlp": 0.01004312,
      "balance_loss_clip": 1.02822042,
      "balance_loss_mlp": 1.00189233,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 0.9766903489581491,
      "language_loss": 0.62455153,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64557171,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 4.084162712097168
    },
    {
      "auxiliary_loss_clip": 0.01192872,
      "auxiliary_loss_mlp": 0.01029018,
      "balance_loss_clip": 1.0591495,
      "balance_loss_mlp": 1.02122188,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 1.7867853700642637,
      "language_loss": 0.81652266,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.83874154,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 2.6024351119995117
    },
    {
      "auxiliary_loss_clip": 0.01199513,
      "auxiliary_loss_mlp": 0.01027472,
      "balance_loss_clip": 0.9445461,
      "balance_loss_mlp": 1.01910341,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 2.009795531299865,
      "language_loss": 0.74475849,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76702833,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.6956582069396973
    },
    {
      "auxiliary_loss_clip": 0.01187108,
      "auxiliary_loss_mlp": 0.01032726,
      "balance_loss_clip": 0.90179604,
      "balance_loss_mlp": 1.02507889,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 1.937398064011089,
      "language_loss": 0.79115695,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.81335521,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 3.614084005355835
    },
    {
      "auxiliary_loss_clip": 0.01184271,
      "auxiliary_loss_mlp": 0.01033896,
      "balance_loss_clip": 0.903965,
      "balance_loss_mlp": 1.02481246,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.664972890561432,
      "language_loss": 0.88514161,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.90732336,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 2.784017324447632
    },
    {
      "auxiliary_loss_clip": 0.01184906,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 0.98095983,
      "balance_loss_mlp": 1.01777887,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 2.1777162972554422,
      "language_loss": 0.68022245,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70232391,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.74819278717041
    },
    {
      "auxiliary_loss_clip": 0.01193607,
      "auxiliary_loss_mlp": 0.01029756,
      "balance_loss_clip": 0.98518836,
      "balance_loss_mlp": 1.0216974,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.477537610499848,
      "language_loss": 0.74973971,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77197337,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.7221155166625977
    },
    {
      "auxiliary_loss_clip": 0.0118092,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 0.86294872,
      "balance_loss_mlp": 1.01672387,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.7378186008405037,
      "language_loss": 0.74846256,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.7705186,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.8326406478881836
    },
    {
      "auxiliary_loss_clip": 0.01189225,
      "auxiliary_loss_mlp": 0.01030652,
      "balance_loss_clip": 0.98098314,
      "balance_loss_mlp": 1.02255201,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 1.941399732231767,
      "language_loss": 0.77328074,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79547954,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 3.679802894592285
    },
    {
      "auxiliary_loss_clip": 0.01196663,
      "auxiliary_loss_mlp": 0.01033576,
      "balance_loss_clip": 1.0193218,
      "balance_loss_mlp": 1.02557743,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 1.8731621170124404,
      "language_loss": 0.79218596,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81448829,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 3.640514850616455
    },
    {
      "auxiliary_loss_clip": 0.0118766,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 0.93981564,
      "balance_loss_mlp": 1.02029014,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.2032743205481435,
      "language_loss": 0.71262538,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.7347821,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.690366268157959
    },
    {
      "auxiliary_loss_clip": 0.01180334,
      "auxiliary_loss_mlp": 0.0103198,
      "balance_loss_clip": 0.97691977,
      "balance_loss_mlp": 1.02350438,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 1.8787713255518135,
      "language_loss": 0.69533646,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.71745962,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.7499961853027344
    },
    {
      "auxiliary_loss_clip": 0.01187088,
      "auxiliary_loss_mlp": 0.01033584,
      "balance_loss_clip": 0.97633457,
      "balance_loss_mlp": 1.0249474,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 2.266729178311097,
      "language_loss": 0.85609031,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.87829709,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.6432952880859375
    },
    {
      "auxiliary_loss_clip": 0.01104405,
      "auxiliary_loss_mlp": 0.01002051,
      "balance_loss_clip": 0.80172396,
      "balance_loss_mlp": 0.99950045,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8799268369430315,
      "language_loss": 0.55917954,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.58024406,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.562037467956543
    },
    {
      "auxiliary_loss_clip": 0.01174426,
      "auxiliary_loss_mlp": 0.0103151,
      "balance_loss_clip": 0.89788288,
      "balance_loss_mlp": 1.02362728,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 1.9399493502680265,
      "language_loss": 0.84644306,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.86850244,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 3.207641363143921
    },
    {
      "auxiliary_loss_clip": 0.01178863,
      "auxiliary_loss_mlp": 0.01031607,
      "balance_loss_clip": 0.85904825,
      "balance_loss_mlp": 1.02397776,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.50652719425137,
      "language_loss": 0.76168287,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.78378755,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.817821502685547
    },
    {
      "auxiliary_loss_clip": 0.01190328,
      "auxiliary_loss_mlp": 0.01025951,
      "balance_loss_clip": 0.9423784,
      "balance_loss_mlp": 1.01851845,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 1.5566932930519266,
      "language_loss": 0.73946583,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.76162863,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 2.782031536102295
    },
    {
      "auxiliary_loss_clip": 0.01189345,
      "auxiliary_loss_mlp": 0.01026246,
      "balance_loss_clip": 0.98240829,
      "balance_loss_mlp": 1.0183661,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 1.6803435942144778,
      "language_loss": 0.82753253,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.84968841,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 3.4188671112060547
    },
    {
      "auxiliary_loss_clip": 0.01185698,
      "auxiliary_loss_mlp": 0.01028165,
      "balance_loss_clip": 0.97900349,
      "balance_loss_mlp": 1.02043986,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 2.025133334061084,
      "language_loss": 0.75833333,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.78047192,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 2.8703441619873047
    },
    {
      "auxiliary_loss_clip": 0.01180068,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 0.86024058,
      "balance_loss_mlp": 1.02159238,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.1341510829655697,
      "language_loss": 0.76357436,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.78567219,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.7995529174804688
    },
    {
      "auxiliary_loss_clip": 0.01195429,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 0.94002277,
      "balance_loss_mlp": 1.02016902,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.7049250614428764,
      "language_loss": 0.81948781,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.84172857,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 2.673051118850708
    },
    {
      "auxiliary_loss_clip": 0.01187073,
      "auxiliary_loss_mlp": 0.01027834,
      "balance_loss_clip": 1.0189141,
      "balance_loss_mlp": 1.02028775,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 1.7175729699417823,
      "language_loss": 0.77944362,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80159271,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.6233532428741455
    },
    {
      "auxiliary_loss_clip": 0.01186046,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.01888955,
      "balance_loss_mlp": 1.02240229,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 2.1040264530927035,
      "language_loss": 0.74149895,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.7636584,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 2.5501067638397217
    },
    {
      "auxiliary_loss_clip": 0.0117952,
      "auxiliary_loss_mlp": 0.01023489,
      "balance_loss_clip": 1.01536298,
      "balance_loss_mlp": 1.01600909,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 1.8312228163131195,
      "language_loss": 0.66910601,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.69113612,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.542752742767334
    },
    {
      "auxiliary_loss_clip": 0.01184349,
      "auxiliary_loss_mlp": 0.01123725,
      "balance_loss_clip": 0.90344644,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.628194336622218,
      "language_loss": 0.69371867,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.71679938,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.688891649246216
    },
    {
      "auxiliary_loss_clip": 0.01190672,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 0.97814989,
      "balance_loss_mlp": 1.01975679,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 2.070116956671189,
      "language_loss": 0.63680857,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.65900254,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 2.794471502304077
    },
    {
      "auxiliary_loss_clip": 0.0119002,
      "auxiliary_loss_mlp": 0.0103285,
      "balance_loss_clip": 1.019333,
      "balance_loss_mlp": 1.02491665,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 4.21600486233936,
      "language_loss": 0.68888342,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.71111214,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 3.8192460536956787
    },
    {
      "auxiliary_loss_clip": 0.0118335,
      "auxiliary_loss_mlp": 0.01029964,
      "balance_loss_clip": 0.97930491,
      "balance_loss_mlp": 1.02189302,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 1.973145540479078,
      "language_loss": 0.7655074,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78764057,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 2.853579044342041
    },
    {
      "auxiliary_loss_clip": 0.01188861,
      "auxiliary_loss_mlp": 0.01023499,
      "balance_loss_clip": 1.0164659,
      "balance_loss_mlp": 1.01536846,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 1.7714235993408396,
      "language_loss": 0.87230432,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89442796,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 3.720607280731201
    },
    {
      "auxiliary_loss_clip": 0.01193961,
      "auxiliary_loss_mlp": 0.0103278,
      "balance_loss_clip": 1.05726957,
      "balance_loss_mlp": 1.02434611,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 2.0868279408509482,
      "language_loss": 0.80209559,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.82436299,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.5926430225372314
    },
    {
      "auxiliary_loss_clip": 0.01195346,
      "auxiliary_loss_mlp": 0.01028065,
      "balance_loss_clip": 1.02011204,
      "balance_loss_mlp": 1.0197506,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 2.5253648998636984,
      "language_loss": 0.85995209,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.88218617,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.602541208267212
    },
    {
      "auxiliary_loss_clip": 0.01193892,
      "auxiliary_loss_mlp": 0.01027156,
      "balance_loss_clip": 1.01595569,
      "balance_loss_mlp": 1.01903224,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.7473309261174979,
      "language_loss": 0.81709868,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.83930916,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.534945011138916
    },
    {
      "auxiliary_loss_clip": 0.0119474,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 0.98107934,
      "balance_loss_mlp": 1.01955628,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.5407588991685244,
      "language_loss": 0.74947977,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.7717064,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.731348752975464
    },
    {
      "auxiliary_loss_clip": 0.01096414,
      "auxiliary_loss_mlp": 0.01004006,
      "balance_loss_clip": 0.98845828,
      "balance_loss_mlp": 1.00165749,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.9792655390816026,
      "language_loss": 0.59241712,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.61342132,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 4.297847509384155
    },
    {
      "auxiliary_loss_clip": 0.01192746,
      "auxiliary_loss_mlp": 0.0103347,
      "balance_loss_clip": 0.9415307,
      "balance_loss_mlp": 1.02489936,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 1.7722824460549318,
      "language_loss": 0.76221216,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78447437,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.8877084255218506
    },
    {
      "auxiliary_loss_clip": 0.01176763,
      "auxiliary_loss_mlp": 0.01027717,
      "balance_loss_clip": 0.97645378,
      "balance_loss_mlp": 1.01922345,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 1.777367658614394,
      "language_loss": 0.6593014,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68134618,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 3.6723198890686035
    },
    {
      "auxiliary_loss_clip": 0.01193885,
      "auxiliary_loss_mlp": 0.01029885,
      "balance_loss_clip": 1.01799297,
      "balance_loss_mlp": 1.02208614,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 1.9172375999342541,
      "language_loss": 0.81296366,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83520138,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.7343382835388184
    },
    {
      "auxiliary_loss_clip": 0.01184963,
      "auxiliary_loss_mlp": 0.01030792,
      "balance_loss_clip": 0.97947264,
      "balance_loss_mlp": 1.02294791,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 3.0292311751725447,
      "language_loss": 0.75014621,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77230376,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 2.6738827228546143
    },
    {
      "auxiliary_loss_clip": 0.01194666,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.05773306,
      "balance_loss_mlp": 1.02410054,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 2.358696797707724,
      "language_loss": 0.84048349,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.86275369,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.74772047996521
    },
    {
      "auxiliary_loss_clip": 0.01175854,
      "auxiliary_loss_mlp": 0.01030768,
      "balance_loss_clip": 0.97373199,
      "balance_loss_mlp": 1.02304292,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.7556697103035925,
      "language_loss": 0.78531939,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.80738562,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.6792619228363037
    },
    {
      "auxiliary_loss_clip": 0.01185679,
      "auxiliary_loss_mlp": 0.01029053,
      "balance_loss_clip": 0.94140387,
      "balance_loss_mlp": 1.02033901,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 1.560805948533513,
      "language_loss": 0.69438922,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.71653658,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.807981252670288
    },
    {
      "auxiliary_loss_clip": 0.01188332,
      "auxiliary_loss_mlp": 0.01031078,
      "balance_loss_clip": 0.9780612,
      "balance_loss_mlp": 1.02341866,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 2.1697432177903524,
      "language_loss": 0.86828423,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89047831,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.7067975997924805
    },
    {
      "auxiliary_loss_clip": 0.0119336,
      "auxiliary_loss_mlp": 0.01026214,
      "balance_loss_clip": 1.01900506,
      "balance_loss_mlp": 1.01743412,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.1474246505380137,
      "language_loss": 0.78309631,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80529201,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 2.6272664070129395
    },
    {
      "auxiliary_loss_clip": 0.01183863,
      "auxiliary_loss_mlp": 0.01029846,
      "balance_loss_clip": 0.97879839,
      "balance_loss_mlp": 1.0217036,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 3.183986919400606,
      "language_loss": 0.83903933,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.86117643,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.6736302375793457
    },
    {
      "auxiliary_loss_clip": 0.01189156,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.05624282,
      "balance_loss_mlp": 1.02027392,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 1.775748456740706,
      "language_loss": 0.79267949,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81485021,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 2.643807888031006
    },
    {
      "auxiliary_loss_clip": 0.01186289,
      "auxiliary_loss_mlp": 0.01023323,
      "balance_loss_clip": 0.98391163,
      "balance_loss_mlp": 1.01576221,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 2.5234743632683343,
      "language_loss": 0.74566078,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.76775694,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.702609062194824
    },
    {
      "auxiliary_loss_clip": 0.01189396,
      "auxiliary_loss_mlp": 0.01032028,
      "balance_loss_clip": 1.05426335,
      "balance_loss_mlp": 1.02456534,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 2.090289000041398,
      "language_loss": 0.8138504,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83606458,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 2.6713809967041016
    },
    {
      "auxiliary_loss_clip": 0.01194628,
      "auxiliary_loss_mlp": 0.01024844,
      "balance_loss_clip": 1.01649165,
      "balance_loss_mlp": 1.01687491,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 1.8416017409570564,
      "language_loss": 0.80225921,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.82445395,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.692624807357788
    },
    {
      "auxiliary_loss_clip": 0.01187618,
      "auxiliary_loss_mlp": 0.01124259,
      "balance_loss_clip": 1.01604605,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.4353344435228976,
      "language_loss": 0.85351175,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.87663043,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.709597110748291
    },
    {
      "auxiliary_loss_clip": 0.01188736,
      "auxiliary_loss_mlp": 0.01026857,
      "balance_loss_clip": 0.94537687,
      "balance_loss_mlp": 1.01869106,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.5302207436259943,
      "language_loss": 0.79280806,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81496394,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 2.84798002243042
    },
    {
      "auxiliary_loss_clip": 0.0118402,
      "auxiliary_loss_mlp": 0.01123966,
      "balance_loss_clip": 0.97560644,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.662444502935159,
      "language_loss": 0.75180715,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.77488697,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.6979987621307373
    },
    {
      "auxiliary_loss_clip": 0.01094099,
      "auxiliary_loss_mlp": 0.01000507,
      "balance_loss_clip": 0.98900139,
      "balance_loss_mlp": 0.99819422,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.8876778193987833,
      "language_loss": 0.65269089,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67363691,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.3371849060058594
    },
    {
      "auxiliary_loss_clip": 0.01180254,
      "auxiliary_loss_mlp": 0.01123771,
      "balance_loss_clip": 0.93684602,
      "balance_loss_mlp": 0.0,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.4682663833820173,
      "language_loss": 0.85336655,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.87640679,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 3.720947742462158
    },
    {
      "auxiliary_loss_clip": 0.01173825,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 0.97506005,
      "balance_loss_mlp": 1.01877069,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 1.96968648194101,
      "language_loss": 0.86144328,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88344777,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 2.784085988998413
    },
    {
      "auxiliary_loss_clip": 0.01191156,
      "auxiliary_loss_mlp": 0.01124837,
      "balance_loss_clip": 1.01770473,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 1.8399448286858455,
      "language_loss": 0.77295601,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.79611599,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 3.6891732215881348
    },
    {
      "auxiliary_loss_clip": 0.01188153,
      "auxiliary_loss_mlp": 0.01028562,
      "balance_loss_clip": 1.01973665,
      "balance_loss_mlp": 1.02098584,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.3347536656200834,
      "language_loss": 0.6973936,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.71956074,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.8524723052978516
    },
    {
      "auxiliary_loss_clip": 0.01192729,
      "auxiliary_loss_mlp": 0.01124247,
      "balance_loss_clip": 1.05548644,
      "balance_loss_mlp": 0.0,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 4.256473890578764,
      "language_loss": 0.72349477,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74666452,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 2.6917896270751953
    },
    {
      "auxiliary_loss_clip": 0.01185236,
      "auxiliary_loss_mlp": 0.01124469,
      "balance_loss_clip": 0.93794,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 2.0811677005704694,
      "language_loss": 0.83330274,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.85639977,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.751016616821289
    },
    {
      "auxiliary_loss_clip": 0.011828,
      "auxiliary_loss_mlp": 0.01025808,
      "balance_loss_clip": 1.01536822,
      "balance_loss_mlp": 1.01771998,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 1.9458192738194615,
      "language_loss": 0.78537881,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.80746484,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.710099697113037
    },
    {
      "auxiliary_loss_clip": 0.01187187,
      "auxiliary_loss_mlp": 0.01025079,
      "balance_loss_clip": 1.05328608,
      "balance_loss_mlp": 1.01761067,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 1.7625726362548548,
      "language_loss": 0.71811271,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.74023539,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 3.7620668411254883
    },
    {
      "auxiliary_loss_clip": 0.01190997,
      "auxiliary_loss_mlp": 0.01026012,
      "balance_loss_clip": 1.01542568,
      "balance_loss_mlp": 1.01837039,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.1345270531803044,
      "language_loss": 0.83946598,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.86163604,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.67633056640625
    },
    {
      "auxiliary_loss_clip": 0.01189901,
      "auxiliary_loss_mlp": 0.01025816,
      "balance_loss_clip": 1.05497098,
      "balance_loss_mlp": 1.01820445,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 2.2561389250278716,
      "language_loss": 0.77795839,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.80011559,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 3.5003535747528076
    },
    {
      "auxiliary_loss_clip": 0.01180355,
      "auxiliary_loss_mlp": 0.01024943,
      "balance_loss_clip": 1.01245546,
      "balance_loss_mlp": 1.01748061,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 1.8012731342457244,
      "language_loss": 0.83486992,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.85692292,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.7257301807403564
    },
    {
      "auxiliary_loss_clip": 0.0118341,
      "auxiliary_loss_mlp": 0.01026101,
      "balance_loss_clip": 0.97549611,
      "balance_loss_mlp": 1.0169456,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 2.161847366436668,
      "language_loss": 0.73651463,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.75860977,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.7461721897125244
    },
    {
      "auxiliary_loss_clip": 0.01185041,
      "auxiliary_loss_mlp": 0.01124317,
      "balance_loss_clip": 0.9759506,
      "balance_loss_mlp": 0.0,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.060881518059898,
      "language_loss": 0.8921988,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.91529238,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.7260191440582275
    },
    {
      "auxiliary_loss_clip": 0.01096072,
      "auxiliary_loss_mlp": 0.01004389,
      "balance_loss_clip": 0.98797643,
      "balance_loss_mlp": 1.00199318,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.879883050359054,
      "language_loss": 0.66181648,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68282104,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.2258365154266357
    },
    {
      "auxiliary_loss_clip": 0.01186009,
      "auxiliary_loss_mlp": 0.010275,
      "balance_loss_clip": 1.01512122,
      "balance_loss_mlp": 1.02008796,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 2.5792220205130745,
      "language_loss": 0.83049846,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85263354,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.631721019744873
    },
    {
      "auxiliary_loss_clip": 0.01186915,
      "auxiliary_loss_mlp": 0.01033814,
      "balance_loss_clip": 1.01442695,
      "balance_loss_mlp": 1.02602386,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 1.6752734470488935,
      "language_loss": 0.75744092,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.77964818,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.7412259578704834
    },
    {
      "auxiliary_loss_clip": 0.01185274,
      "auxiliary_loss_mlp": 0.01027559,
      "balance_loss_clip": 1.05290627,
      "balance_loss_mlp": 1.01970267,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.5055794001881408,
      "language_loss": 0.72414863,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.74627697,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 2.6237001419067383
    },
    {
      "auxiliary_loss_clip": 0.01179959,
      "auxiliary_loss_mlp": 0.01026726,
      "balance_loss_clip": 0.93542159,
      "balance_loss_mlp": 1.01840568,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 2.38885432817769,
      "language_loss": 0.81527102,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83733785,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.8329670429229736
    },
    {
      "auxiliary_loss_clip": 0.0119118,
      "auxiliary_loss_mlp": 0.01026001,
      "balance_loss_clip": 1.01921916,
      "balance_loss_mlp": 1.01806188,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.2085910305285634,
      "language_loss": 0.70982331,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.73199511,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 2.6462676525115967
    },
    {
      "auxiliary_loss_clip": 0.01190107,
      "auxiliary_loss_mlp": 0.01025072,
      "balance_loss_clip": 1.05485964,
      "balance_loss_mlp": 1.01660275,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.5650261831707777,
      "language_loss": 0.78984714,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.81199896,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.6925888061523438
    },
    {
      "auxiliary_loss_clip": 0.01086699,
      "auxiliary_loss_mlp": 0.01002102,
      "balance_loss_clip": 0.98502946,
      "balance_loss_mlp": 0.9998371,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8624332736926908,
      "language_loss": 0.62924218,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.65013021,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.286583662033081
    },
    {
      "auxiliary_loss_clip": 0.0118803,
      "auxiliary_loss_mlp": 0.01025615,
      "balance_loss_clip": 1.05453944,
      "balance_loss_mlp": 1.01731241,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 1.813709126475665,
      "language_loss": 0.73947012,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76160657,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.7270121574401855
    },
    {
      "auxiliary_loss_clip": 0.01188982,
      "auxiliary_loss_mlp": 0.01025056,
      "balance_loss_clip": 1.05488002,
      "balance_loss_mlp": 1.01731396,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.6532327782384963,
      "language_loss": 0.76109731,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.7832377,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 2.6522912979125977
    },
    {
      "auxiliary_loss_clip": 0.01177118,
      "auxiliary_loss_mlp": 0.01028676,
      "balance_loss_clip": 1.01219106,
      "balance_loss_mlp": 1.02039123,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 1.884634817819773,
      "language_loss": 0.73343128,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75548923,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.677262783050537
    },
    {
      "auxiliary_loss_clip": 0.01187223,
      "auxiliary_loss_mlp": 0.01031751,
      "balance_loss_clip": 1.0158354,
      "balance_loss_mlp": 1.02403784,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 2.0245396395211377,
      "language_loss": 0.76657599,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78876567,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.708516836166382
    },
    {
      "auxiliary_loss_clip": 0.0118872,
      "auxiliary_loss_mlp": 0.01032215,
      "balance_loss_clip": 1.01705945,
      "balance_loss_mlp": 1.02484226,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.5203263357004682,
      "language_loss": 0.73828197,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.76049125,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.672978639602661
    },
    {
      "auxiliary_loss_clip": 0.01096336,
      "auxiliary_loss_mlp": 0.0100157,
      "balance_loss_clip": 0.94925946,
      "balance_loss_mlp": 0.99923342,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.9030612285150593,
      "language_loss": 0.62801313,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64899224,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 4.320513963699341
    },
    {
      "auxiliary_loss_clip": 0.0118712,
      "auxiliary_loss_mlp": 0.01027264,
      "balance_loss_clip": 1.01528203,
      "balance_loss_mlp": 1.01926446,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 1.8355817044659142,
      "language_loss": 0.78499299,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.80713683,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 2.7179958820343018
    },
    {
      "auxiliary_loss_clip": 0.01178689,
      "auxiliary_loss_mlp": 0.01029086,
      "balance_loss_clip": 0.97814965,
      "balance_loss_mlp": 1.02050877,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 1.6879413097163805,
      "language_loss": 0.79148501,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81356275,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 3.6406211853027344
    },
    {
      "auxiliary_loss_clip": 0.01187277,
      "auxiliary_loss_mlp": 0.01028874,
      "balance_loss_clip": 0.89940202,
      "balance_loss_mlp": 1.02076173,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 3.390824287117582,
      "language_loss": 0.83333576,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85549724,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.8253984451293945
    },
    {
      "auxiliary_loss_clip": 0.01101363,
      "auxiliary_loss_mlp": 0.00999924,
      "balance_loss_clip": 0.8735671,
      "balance_loss_mlp": 0.99730128,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7702844748202228,
      "language_loss": 0.53038818,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55140102,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.4243438243865967
    },
    {
      "auxiliary_loss_clip": 0.01178353,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 0.97546995,
      "balance_loss_mlp": 1.01746869,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 1.7742768115252951,
      "language_loss": 0.82476592,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.8468014,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 2.987236261367798
    },
    {
      "auxiliary_loss_clip": 0.01086403,
      "auxiliary_loss_mlp": 0.01003731,
      "balance_loss_clip": 0.79469085,
      "balance_loss_mlp": 1.00115609,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.8263135531820458,
      "language_loss": 0.59334046,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61424178,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.4625842571258545
    },
    {
      "auxiliary_loss_clip": 0.01167516,
      "auxiliary_loss_mlp": 0.01024969,
      "balance_loss_clip": 0.97088999,
      "balance_loss_mlp": 1.01648724,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.5196174768477186,
      "language_loss": 0.78480196,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.80672681,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 4.605800628662109
    },
    {
      "auxiliary_loss_clip": 0.01171722,
      "auxiliary_loss_mlp": 0.01032376,
      "balance_loss_clip": 0.97302949,
      "balance_loss_mlp": 1.02466345,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 2.122882032124826,
      "language_loss": 0.83580601,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.85784698,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.72011399269104
    },
    {
      "auxiliary_loss_clip": 0.01095142,
      "auxiliary_loss_mlp": 0.00999938,
      "balance_loss_clip": 0.87541449,
      "balance_loss_mlp": 0.99749464,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7241888066833814,
      "language_loss": 0.53950995,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.56046075,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 4.195053815841675
    },
    {
      "auxiliary_loss_clip": 0.0118762,
      "auxiliary_loss_mlp": 0.01028797,
      "balance_loss_clip": 0.97872341,
      "balance_loss_mlp": 1.02061975,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 3.568520692230349,
      "language_loss": 0.7846365,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.8068006,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.747774600982666
    },
    {
      "auxiliary_loss_clip": 0.01183486,
      "auxiliary_loss_mlp": 0.01124272,
      "balance_loss_clip": 0.93711448,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 1.9698785361593676,
      "language_loss": 0.78281617,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.80589378,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.7431743144989014
    },
    {
      "auxiliary_loss_clip": 0.01088487,
      "auxiliary_loss_mlp": 0.01006355,
      "balance_loss_clip": 1.02266288,
      "balance_loss_mlp": 1.00394714,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.8136543830857919,
      "language_loss": 0.59237725,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.61332566,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 3.2574870586395264
    },
    {
      "auxiliary_loss_clip": 0.01188112,
      "auxiliary_loss_mlp": 0.01024191,
      "balance_loss_clip": 1.0548923,
      "balance_loss_mlp": 1.01660919,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.0157502351508323,
      "language_loss": 0.68584716,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.7079702,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.5962326526641846
    },
    {
      "auxiliary_loss_clip": 0.01189898,
      "auxiliary_loss_mlp": 0.01035972,
      "balance_loss_clip": 1.01693726,
      "balance_loss_mlp": 1.02780032,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 1.7919233238813588,
      "language_loss": 0.8238135,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.8460722,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.676645278930664
    },
    {
      "auxiliary_loss_clip": 0.01171663,
      "auxiliary_loss_mlp": 0.01030972,
      "balance_loss_clip": 0.89965457,
      "balance_loss_mlp": 1.02250814,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 1.7769012318073396,
      "language_loss": 0.76876897,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.79079533,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.8385722637176514
    },
    {
      "auxiliary_loss_clip": 0.01174556,
      "auxiliary_loss_mlp": 0.01025898,
      "balance_loss_clip": 0.85998464,
      "balance_loss_mlp": 1.0167371,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 1.8043779827002198,
      "language_loss": 0.78375775,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.80576229,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.813248872756958
    },
    {
      "auxiliary_loss_clip": 0.01178253,
      "auxiliary_loss_mlp": 0.01027387,
      "balance_loss_clip": 0.938398,
      "balance_loss_mlp": 1.01937056,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.293173589848934,
      "language_loss": 0.66559726,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.68765366,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 2.8560667037963867
    },
    {
      "auxiliary_loss_clip": 0.01189023,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.0164938,
      "balance_loss_mlp": 1.01870942,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 3.504636592440565,
      "language_loss": 0.76272744,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.78488588,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 2.633114814758301
    },
    {
      "auxiliary_loss_clip": 0.01183118,
      "auxiliary_loss_mlp": 0.01034744,
      "balance_loss_clip": 0.97809273,
      "balance_loss_mlp": 1.02582669,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 2.643502136234976,
      "language_loss": 0.79824907,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82042766,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.9048571586608887
    },
    {
      "auxiliary_loss_clip": 0.01181992,
      "auxiliary_loss_mlp": 0.01027013,
      "balance_loss_clip": 0.93817794,
      "balance_loss_mlp": 1.0181675,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 1.8798652983849298,
      "language_loss": 0.76203549,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78412551,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.702089309692383
    },
    {
      "auxiliary_loss_clip": 0.01193075,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 0.97770834,
      "balance_loss_mlp": 1.02014685,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 2.095289321818517,
      "language_loss": 0.74489582,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76710939,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.6747171878814697
    },
    {
      "auxiliary_loss_clip": 0.01180794,
      "auxiliary_loss_mlp": 0.0112471,
      "balance_loss_clip": 0.93600428,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 1.7964861925520441,
      "language_loss": 0.80317056,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.82622564,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 2.8168234825134277
    },
    {
      "auxiliary_loss_clip": 0.01189082,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 0.98094237,
      "balance_loss_mlp": 1.01939535,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.6451811290945086,
      "language_loss": 0.73855114,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76071417,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.687913179397583
    },
    {
      "auxiliary_loss_clip": 0.01188252,
      "auxiliary_loss_mlp": 0.01032016,
      "balance_loss_clip": 1.0192064,
      "balance_loss_mlp": 1.02395201,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.6907552328330453,
      "language_loss": 0.83803231,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.86023498,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.665295362472534
    },
    {
      "auxiliary_loss_clip": 0.01181118,
      "auxiliary_loss_mlp": 0.0102796,
      "balance_loss_clip": 0.93658459,
      "balance_loss_mlp": 1.01946616,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 1.6956745393389605,
      "language_loss": 0.71570605,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73779678,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.820096731185913
    },
    {
      "auxiliary_loss_clip": 0.01188604,
      "auxiliary_loss_mlp": 0.01029279,
      "balance_loss_clip": 0.97543383,
      "balance_loss_mlp": 1.02095783,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 1.8864130174019806,
      "language_loss": 0.7464776,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.76865649,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 4.133085489273071
    },
    {
      "auxiliary_loss_clip": 0.01189992,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 0.93927586,
      "balance_loss_mlp": 1.02060485,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 2.84785174638537,
      "language_loss": 0.79029471,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.81248516,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 2.8135123252868652
    },
    {
      "auxiliary_loss_clip": 0.01184148,
      "auxiliary_loss_mlp": 0.01036573,
      "balance_loss_clip": 0.82158923,
      "balance_loss_mlp": 1.02787066,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 2.965503672438945,
      "language_loss": 0.76041484,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.78262198,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 3.886563301086426
    },
    {
      "auxiliary_loss_clip": 0.01186464,
      "auxiliary_loss_mlp": 0.01027315,
      "balance_loss_clip": 1.0184617,
      "balance_loss_mlp": 1.01928663,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.5152575835402569,
      "language_loss": 0.78245234,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.80459011,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.989163637161255
    },
    {
      "auxiliary_loss_clip": 0.0118389,
      "auxiliary_loss_mlp": 0.01033602,
      "balance_loss_clip": 0.86271095,
      "balance_loss_mlp": 1.02509046,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 1.7555028459046753,
      "language_loss": 0.80944276,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.83161759,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.8305187225341797
    },
    {
      "auxiliary_loss_clip": 0.01188968,
      "auxiliary_loss_mlp": 0.01028876,
      "balance_loss_clip": 0.93968582,
      "balance_loss_mlp": 1.02097845,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 1.7600587274916155,
      "language_loss": 0.79195476,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81413317,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 2.8433892726898193
    },
    {
      "auxiliary_loss_clip": 0.01191782,
      "auxiliary_loss_mlp": 0.01027155,
      "balance_loss_clip": 1.05597901,
      "balance_loss_mlp": 1.01841152,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.241016450093393,
      "language_loss": 0.76801753,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.79020691,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.6364293098449707
    },
    {
      "auxiliary_loss_clip": 0.01176681,
      "auxiliary_loss_mlp": 0.01023607,
      "balance_loss_clip": 0.97555476,
      "balance_loss_mlp": 1.01520276,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 3.004285210027862,
      "language_loss": 0.79128647,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.8132894,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 3.9481306076049805
    },
    {
      "auxiliary_loss_clip": 0.01185673,
      "auxiliary_loss_mlp": 0.01030701,
      "balance_loss_clip": 0.98013192,
      "balance_loss_mlp": 1.02260685,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 1.950897197756079,
      "language_loss": 0.65220881,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.67437255,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.7649476528167725
    },
    {
      "auxiliary_loss_clip": 0.01184991,
      "auxiliary_loss_mlp": 0.01123756,
      "balance_loss_clip": 1.05311942,
      "balance_loss_mlp": 0.0,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 1.4715661953335895,
      "language_loss": 0.79867113,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.82175863,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 3.866159200668335
    },
    {
      "auxiliary_loss_clip": 0.01187212,
      "auxiliary_loss_mlp": 0.0103495,
      "balance_loss_clip": 1.01957154,
      "balance_loss_mlp": 1.02684951,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.0693629509356244,
      "language_loss": 0.8442874,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86650908,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.6857118606567383
    },
    {
      "auxiliary_loss_clip": 0.01189108,
      "auxiliary_loss_mlp": 0.01031571,
      "balance_loss_clip": 1.05343032,
      "balance_loss_mlp": 1.02317858,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 1.6924394182770377,
      "language_loss": 0.73062563,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75283247,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.712351083755493
    },
    {
      "auxiliary_loss_clip": 0.01179974,
      "auxiliary_loss_mlp": 0.0103093,
      "balance_loss_clip": 1.01342916,
      "balance_loss_mlp": 1.02275848,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 2.3018997307637794,
      "language_loss": 0.80687571,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.82898474,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.730107545852661
    },
    {
      "auxiliary_loss_clip": 0.01187111,
      "auxiliary_loss_mlp": 0.01029119,
      "balance_loss_clip": 0.97923869,
      "balance_loss_mlp": 1.02107882,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 2.32076706972064,
      "language_loss": 0.75132906,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.77349132,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 2.72652006149292
    },
    {
      "auxiliary_loss_clip": 0.01190842,
      "auxiliary_loss_mlp": 0.01030507,
      "balance_loss_clip": 1.0146879,
      "balance_loss_mlp": 1.02237654,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 1.8171953407504602,
      "language_loss": 0.65414405,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67635751,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.6925618648529053
    },
    {
      "auxiliary_loss_clip": 0.0117796,
      "auxiliary_loss_mlp": 0.01035759,
      "balance_loss_clip": 0.90065885,
      "balance_loss_mlp": 1.02756369,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 1.9826081028998828,
      "language_loss": 0.79317641,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81531358,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.9608187675476074
    },
    {
      "auxiliary_loss_clip": 0.01191162,
      "auxiliary_loss_mlp": 0.01029323,
      "balance_loss_clip": 1.05709827,
      "balance_loss_mlp": 1.0205071,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 1.7911810541851854,
      "language_loss": 0.80201226,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.82421714,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.623842716217041
    },
    {
      "auxiliary_loss_clip": 0.01186024,
      "auxiliary_loss_mlp": 0.01023826,
      "balance_loss_clip": 1.01897812,
      "balance_loss_mlp": 1.01568449,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 1.89728769238286,
      "language_loss": 0.80101901,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82311749,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.760883331298828
    },
    {
      "auxiliary_loss_clip": 0.01185872,
      "auxiliary_loss_mlp": 0.01024942,
      "balance_loss_clip": 1.01641631,
      "balance_loss_mlp": 1.01730704,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 1.8881072203699492,
      "language_loss": 0.84210348,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.86421162,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 2.7281532287597656
    },
    {
      "auxiliary_loss_clip": 0.0119055,
      "auxiliary_loss_mlp": 0.01033876,
      "balance_loss_clip": 1.01866519,
      "balance_loss_mlp": 1.02631259,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 1.838759244959022,
      "language_loss": 0.79217851,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81442279,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.676762104034424
    },
    {
      "auxiliary_loss_clip": 0.0118022,
      "auxiliary_loss_mlp": 0.01027943,
      "balance_loss_clip": 0.97769916,
      "balance_loss_mlp": 1.01996779,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 1.9012797855220147,
      "language_loss": 0.72376966,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.74585128,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 2.7475123405456543
    },
    {
      "auxiliary_loss_clip": 0.01189621,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 0.98082423,
      "balance_loss_mlp": 1.01994896,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.0572093988994924,
      "language_loss": 0.64013219,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66231567,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.730544090270996
    },
    {
      "auxiliary_loss_clip": 0.01195876,
      "auxiliary_loss_mlp": 0.0102647,
      "balance_loss_clip": 0.94288403,
      "balance_loss_mlp": 1.01840568,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 2.6581600296417425,
      "language_loss": 0.77728462,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.79950809,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 2.826087236404419
    },
    {
      "auxiliary_loss_clip": 0.01188762,
      "auxiliary_loss_mlp": 0.01029236,
      "balance_loss_clip": 1.05468249,
      "balance_loss_mlp": 1.02019417,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.481070133697372,
      "language_loss": 0.69927722,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.72145724,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 2.7416799068450928
    },
    {
      "auxiliary_loss_clip": 0.01180403,
      "auxiliary_loss_mlp": 0.01024141,
      "balance_loss_clip": 0.94065094,
      "balance_loss_mlp": 1.01595163,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.372435307937371,
      "language_loss": 0.78846323,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.81050861,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.764953136444092
    },
    {
      "auxiliary_loss_clip": 0.01192271,
      "auxiliary_loss_mlp": 0.01034359,
      "balance_loss_clip": 0.9811334,
      "balance_loss_mlp": 1.02665198,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 2.717545717469259,
      "language_loss": 0.67337525,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69564158,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.722651481628418
    },
    {
      "auxiliary_loss_clip": 0.01186527,
      "auxiliary_loss_mlp": 0.01028096,
      "balance_loss_clip": 0.97668499,
      "balance_loss_mlp": 1.02024055,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 2.2060311210306605,
      "language_loss": 0.69255984,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.71470612,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 3.7683727741241455
    },
    {
      "auxiliary_loss_clip": 0.01189095,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.01575971,
      "balance_loss_mlp": 1.01801896,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 1.8135340345078441,
      "language_loss": 0.79541051,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.81756532,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 2.8446602821350098
    },
    {
      "auxiliary_loss_clip": 0.01194415,
      "auxiliary_loss_mlp": 0.0103599,
      "balance_loss_clip": 0.97804368,
      "balance_loss_mlp": 1.02738261,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.599435743829535,
      "language_loss": 0.73704958,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.75935352,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 2.7623229026794434
    },
    {
      "auxiliary_loss_clip": 0.01182986,
      "auxiliary_loss_mlp": 0.01028756,
      "balance_loss_clip": 0.93973529,
      "balance_loss_mlp": 1.01990485,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 21.49601175696693,
      "language_loss": 0.8958075,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.91792488,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 3.711980104446411
    },
    {
      "auxiliary_loss_clip": 0.01186911,
      "auxiliary_loss_mlp": 0.01026794,
      "balance_loss_clip": 1.01932228,
      "balance_loss_mlp": 1.01898599,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 1.7105148005664623,
      "language_loss": 0.78968292,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.81181991,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.632242202758789
    },
    {
      "auxiliary_loss_clip": 0.01179567,
      "auxiliary_loss_mlp": 0.01034249,
      "balance_loss_clip": 0.93670273,
      "balance_loss_mlp": 1.02558887,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 1.9707178286085192,
      "language_loss": 0.81306159,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.83519971,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.771085023880005
    },
    {
      "auxiliary_loss_clip": 0.01182574,
      "auxiliary_loss_mlp": 0.01031997,
      "balance_loss_clip": 0.97710103,
      "balance_loss_mlp": 1.02377748,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 1.7163946296361312,
      "language_loss": 0.73931181,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.76145756,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 2.785839796066284
    },
    {
      "auxiliary_loss_clip": 0.01190809,
      "auxiliary_loss_mlp": 0.01029865,
      "balance_loss_clip": 1.01482582,
      "balance_loss_mlp": 1.02137721,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.0910714843992038,
      "language_loss": 0.76920742,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.79141414,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 3.89057993888855
    },
    {
      "auxiliary_loss_clip": 0.01191974,
      "auxiliary_loss_mlp": 0.01031915,
      "balance_loss_clip": 1.01703393,
      "balance_loss_mlp": 1.02384424,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 2.4061407277844333,
      "language_loss": 0.82080781,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84304667,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.71869158744812
    },
    {
      "auxiliary_loss_clip": 0.0118726,
      "auxiliary_loss_mlp": 0.01032954,
      "balance_loss_clip": 1.01725805,
      "balance_loss_mlp": 1.02465105,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.184754659349082,
      "language_loss": 0.80208409,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.82428622,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 3.6479368209838867
    },
    {
      "auxiliary_loss_clip": 0.01183427,
      "auxiliary_loss_mlp": 0.0103348,
      "balance_loss_clip": 1.01245904,
      "balance_loss_mlp": 1.02540052,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 1.6842368257772715,
      "language_loss": 0.7531842,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77535331,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.6602609157562256
    },
    {
      "auxiliary_loss_clip": 0.01188596,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 0.94113147,
      "balance_loss_mlp": 1.01871717,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 1.9307021704777538,
      "language_loss": 0.74826753,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.7704258,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.7853682041168213
    },
    {
      "auxiliary_loss_clip": 0.01112611,
      "auxiliary_loss_mlp": 0.01119491,
      "balance_loss_clip": 0.96389067,
      "balance_loss_mlp": 0.0,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.832409822098683,
      "language_loss": 0.62812489,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.65044588,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.36919903755188
    },
    {
      "auxiliary_loss_clip": 0.01170476,
      "auxiliary_loss_mlp": 0.01025695,
      "balance_loss_clip": 0.93309867,
      "balance_loss_mlp": 1.017344,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 1.9276678175016155,
      "language_loss": 0.79100871,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.81297046,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.740190267562866
    },
    {
      "auxiliary_loss_clip": 0.01189949,
      "auxiliary_loss_mlp": 0.01027076,
      "balance_loss_clip": 1.05562258,
      "balance_loss_mlp": 1.01843929,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 1.7256239283319057,
      "language_loss": 0.79201293,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81418318,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 2.701134204864502
    },
    {
      "auxiliary_loss_clip": 0.01174995,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 0.97363973,
      "balance_loss_mlp": 1.02267802,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 1.7642937027482246,
      "language_loss": 0.64012432,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.66219169,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.701906681060791
    },
    {
      "auxiliary_loss_clip": 0.01183492,
      "auxiliary_loss_mlp": 0.010269,
      "balance_loss_clip": 1.01221752,
      "balance_loss_mlp": 1.01893687,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 1.722820097238357,
      "language_loss": 0.85631204,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87841594,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.7006590366363525
    },
    {
      "auxiliary_loss_clip": 0.01189731,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.05524862,
      "balance_loss_mlp": 1.0178988,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 3.2756999073342348,
      "language_loss": 0.84076393,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86292279,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.6840436458587646
    },
    {
      "auxiliary_loss_clip": 0.01098221,
      "auxiliary_loss_mlp": 0.01013723,
      "balance_loss_clip": 0.8775574,
      "balance_loss_mlp": 1.01126754,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.8302810064182353,
      "language_loss": 0.57401901,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59513849,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 3.358896255493164
    },
    {
      "auxiliary_loss_clip": 0.01192266,
      "auxiliary_loss_mlp": 0.01029782,
      "balance_loss_clip": 1.05656016,
      "balance_loss_mlp": 1.02185488,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 2.2404855980746126,
      "language_loss": 0.81974632,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.84196675,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.830672264099121
    },
    {
      "auxiliary_loss_clip": 0.01181154,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 0.97864765,
      "balance_loss_mlp": 1.01870573,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.558478278309921,
      "language_loss": 0.7351563,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.75723368,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 2.828996419906616
    },
    {
      "auxiliary_loss_clip": 0.01189634,
      "auxiliary_loss_mlp": 0.01031091,
      "balance_loss_clip": 1.01615191,
      "balance_loss_mlp": 1.02336621,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 1.6061375620619993,
      "language_loss": 0.65657854,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.6787858,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.763765335083008
    },
    {
      "auxiliary_loss_clip": 0.01183719,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.01648545,
      "balance_loss_mlp": 1.0186975,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.8235303888072636,
      "language_loss": 0.82171297,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84381169,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.767437696456909
    },
    {
      "auxiliary_loss_clip": 0.01187598,
      "auxiliary_loss_mlp": 0.01027435,
      "balance_loss_clip": 0.93818331,
      "balance_loss_mlp": 1.01943564,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 1.876179053758285,
      "language_loss": 0.82619524,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84834552,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 2.8168582916259766
    },
    {
      "auxiliary_loss_clip": 0.01183312,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 0.93853396,
      "balance_loss_mlp": 1.02225494,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 1.4091360017151375,
      "language_loss": 0.6830523,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.70518512,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.7555437088012695
    },
    {
      "auxiliary_loss_clip": 0.01183219,
      "auxiliary_loss_mlp": 0.01028181,
      "balance_loss_clip": 0.97626567,
      "balance_loss_mlp": 1.01978898,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 2.228077785058058,
      "language_loss": 0.74654245,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76865649,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.7437186241149902
    },
    {
      "auxiliary_loss_clip": 0.0117652,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 0.9367854,
      "balance_loss_mlp": 1.0189116,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.5739851519273946,
      "language_loss": 0.80792826,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.82996064,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 3.7577106952667236
    },
    {
      "auxiliary_loss_clip": 0.01183817,
      "auxiliary_loss_mlp": 0.01124003,
      "balance_loss_clip": 1.01790333,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 2.154868501866276,
      "language_loss": 0.90480053,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.92787874,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 2.7320518493652344
    },
    {
      "auxiliary_loss_clip": 0.01188508,
      "auxiliary_loss_mlp": 0.01026234,
      "balance_loss_clip": 0.94096375,
      "balance_loss_mlp": 1.0183183,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 1.607713957090204,
      "language_loss": 0.7666719,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.78881925,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 3.6696553230285645
    },
    {
      "auxiliary_loss_clip": 0.01095566,
      "auxiliary_loss_mlp": 0.01005812,
      "balance_loss_clip": 1.02994466,
      "balance_loss_mlp": 1.00355935,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7308255094502573,
      "language_loss": 0.58800995,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.60902369,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.304527521133423
    },
    {
      "auxiliary_loss_clip": 0.01180977,
      "auxiliary_loss_mlp": 0.01024135,
      "balance_loss_clip": 1.0513165,
      "balance_loss_mlp": 1.01641583,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.4951291931924562,
      "language_loss": 0.69841146,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.72046262,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.7032618522644043
    },
    {
      "auxiliary_loss_clip": 0.01178862,
      "auxiliary_loss_mlp": 0.01029865,
      "balance_loss_clip": 1.01380241,
      "balance_loss_mlp": 1.02163339,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 2.514918020559133,
      "language_loss": 0.83749849,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.85958582,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.663862466812134
    },
    {
      "auxiliary_loss_clip": 0.01179977,
      "auxiliary_loss_mlp": 0.01124326,
      "balance_loss_clip": 0.9740752,
      "balance_loss_mlp": 0.0,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.6181211382678755,
      "language_loss": 0.73131341,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.7543565,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.763366460800171
    },
    {
      "auxiliary_loss_clip": 0.01184387,
      "auxiliary_loss_mlp": 0.0102192,
      "balance_loss_clip": 1.01903224,
      "balance_loss_mlp": 1.01429021,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 1.8468322275829887,
      "language_loss": 0.8115325,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83359551,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 3.7535085678100586
    },
    {
      "auxiliary_loss_clip": 0.01184112,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 0.97737396,
      "balance_loss_mlp": 1.02092004,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.8253396109424747,
      "language_loss": 0.8269462,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.84907317,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.78688645362854
    },
    {
      "auxiliary_loss_clip": 0.01175413,
      "auxiliary_loss_mlp": 0.01024873,
      "balance_loss_clip": 1.01377892,
      "balance_loss_mlp": 1.01674247,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 1.8445058750215084,
      "language_loss": 0.81759578,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.83959866,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 3.675887107849121
    },
    {
      "auxiliary_loss_clip": 0.0117071,
      "auxiliary_loss_mlp": 0.01028049,
      "balance_loss_clip": 0.86239403,
      "balance_loss_mlp": 1.01995504,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 1.7046074720610764,
      "language_loss": 0.73609614,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.75808376,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.763711929321289
    },
    {
      "auxiliary_loss_clip": 0.01186661,
      "auxiliary_loss_mlp": 0.01123552,
      "balance_loss_clip": 1.01763439,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 1.849461760474183,
      "language_loss": 0.82198608,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.84508824,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.602813243865967
    },
    {
      "auxiliary_loss_clip": 0.01183251,
      "auxiliary_loss_mlp": 0.01029501,
      "balance_loss_clip": 0.97408801,
      "balance_loss_mlp": 1.02166867,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 1.7586196330549413,
      "language_loss": 0.79735816,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.81948566,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.7489404678344727
    },
    {
      "auxiliary_loss_clip": 0.01177837,
      "auxiliary_loss_mlp": 0.01029418,
      "balance_loss_clip": 0.82740116,
      "balance_loss_mlp": 1.02113271,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 2.423790069492652,
      "language_loss": 0.84044045,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86251295,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.844003677368164
    },
    {
      "auxiliary_loss_clip": 0.01099907,
      "auxiliary_loss_mlp": 0.01000699,
      "balance_loss_clip": 0.95316505,
      "balance_loss_mlp": 0.99843448,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9679550067174092,
      "language_loss": 0.72184753,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74285364,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.2908663749694824
    },
    {
      "auxiliary_loss_clip": 0.01184312,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 0.89985716,
      "balance_loss_mlp": 1.01870775,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 1.8415505654874702,
      "language_loss": 0.81649423,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.83860219,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 2.789083242416382
    },
    {
      "auxiliary_loss_clip": 0.01102892,
      "auxiliary_loss_mlp": 0.01011721,
      "balance_loss_clip": 0.96512967,
      "balance_loss_mlp": 1.00953913,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7688424752823143,
      "language_loss": 0.54930007,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.57044619,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 3.225407361984253
    },
    {
      "auxiliary_loss_clip": 0.01186208,
      "auxiliary_loss_mlp": 0.01028269,
      "balance_loss_clip": 1.01494503,
      "balance_loss_mlp": 1.02028203,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.09009621630617,
      "language_loss": 0.77350473,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.79564953,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.7008984088897705
    },
    {
      "auxiliary_loss_clip": 0.01182854,
      "auxiliary_loss_mlp": 0.01031789,
      "balance_loss_clip": 1.01795423,
      "balance_loss_mlp": 1.02387381,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 1.8118415347102772,
      "language_loss": 0.88280869,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.90495515,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.727943181991577
    },
    {
      "auxiliary_loss_clip": 0.01189489,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.05617642,
      "balance_loss_mlp": 1.02282715,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 2.855150416191539,
      "language_loss": 0.93541265,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.95761311,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 2.739384889602661
    },
    {
      "auxiliary_loss_clip": 0.01186944,
      "auxiliary_loss_mlp": 0.01032206,
      "balance_loss_clip": 1.01746249,
      "balance_loss_mlp": 1.02495193,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 4.565914559556942,
      "language_loss": 0.7312144,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75340581,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 2.662269115447998
    },
    {
      "auxiliary_loss_clip": 0.01178843,
      "auxiliary_loss_mlp": 0.01028427,
      "balance_loss_clip": 0.93678606,
      "balance_loss_mlp": 1.02021933,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 2.8222482662010826,
      "language_loss": 0.85057944,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87265217,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.7752909660339355
    },
    {
      "auxiliary_loss_clip": 0.01179896,
      "auxiliary_loss_mlp": 0.01123483,
      "balance_loss_clip": 0.97969902,
      "balance_loss_mlp": 0.0,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.5964906008498059,
      "language_loss": 0.84654772,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.86958158,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.694809913635254
    },
    {
      "auxiliary_loss_clip": 0.01179834,
      "auxiliary_loss_mlp": 0.01027338,
      "balance_loss_clip": 0.89677572,
      "balance_loss_mlp": 1.01910639,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 3.556696744787129,
      "language_loss": 0.79624879,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.81832045,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 2.9223415851593018
    },
    {
      "auxiliary_loss_clip": 0.0116983,
      "auxiliary_loss_mlp": 0.01024731,
      "balance_loss_clip": 0.93733102,
      "balance_loss_mlp": 1.01670241,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 1.6770855256298571,
      "language_loss": 0.89566708,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91761273,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 2.841647148132324
    },
    {
      "auxiliary_loss_clip": 0.01185359,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.05152941,
      "balance_loss_mlp": 1.02532434,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 7.248791152136381,
      "language_loss": 0.80841726,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.83060491,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.653513193130493
    },
    {
      "auxiliary_loss_clip": 0.01185456,
      "auxiliary_loss_mlp": 0.01029781,
      "balance_loss_clip": 0.94197345,
      "balance_loss_mlp": 1.02239871,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 1.9899274232279656,
      "language_loss": 0.74318731,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76533967,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 3.6862852573394775
    },
    {
      "auxiliary_loss_clip": 0.01187796,
      "auxiliary_loss_mlp": 0.01030348,
      "balance_loss_clip": 1.01695263,
      "balance_loss_mlp": 1.02263486,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 1.890475176010539,
      "language_loss": 0.87297934,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89516079,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 2.6700541973114014
    },
    {
      "auxiliary_loss_clip": 0.01189112,
      "auxiliary_loss_mlp": 0.01029089,
      "balance_loss_clip": 0.94002223,
      "balance_loss_mlp": 1.02137041,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 1.5815003144807638,
      "language_loss": 0.88215208,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.90433407,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 3.817883014678955
    },
    {
      "auxiliary_loss_clip": 0.01188208,
      "auxiliary_loss_mlp": 0.01030353,
      "balance_loss_clip": 0.97721976,
      "balance_loss_mlp": 1.02179432,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 2.520955338745875,
      "language_loss": 0.84757984,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.86976552,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 2.7012035846710205
    },
    {
      "auxiliary_loss_clip": 0.01176391,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 0.9003576,
      "balance_loss_mlp": 1.01882041,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 2.0434440641596985,
      "language_loss": 0.74331057,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.76534188,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.7494781017303467
    },
    {
      "auxiliary_loss_clip": 0.01085389,
      "auxiliary_loss_mlp": 0.01002625,
      "balance_loss_clip": 0.98464906,
      "balance_loss_mlp": 1.00043154,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.8010458440255932,
      "language_loss": 0.6213432,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.64222336,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.232682228088379
    },
    {
      "auxiliary_loss_clip": 0.01186234,
      "auxiliary_loss_mlp": 0.0103056,
      "balance_loss_clip": 1.05441511,
      "balance_loss_mlp": 1.02280569,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 1.8957259989162598,
      "language_loss": 0.80950546,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83167344,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.606093406677246
    },
    {
      "auxiliary_loss_clip": 0.01186796,
      "auxiliary_loss_mlp": 0.01034767,
      "balance_loss_clip": 1.01685238,
      "balance_loss_mlp": 1.02646375,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 1.9104338663672158,
      "language_loss": 0.76251894,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78473461,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 3.5986874103546143
    },
    {
      "auxiliary_loss_clip": 0.01179995,
      "auxiliary_loss_mlp": 0.01030794,
      "balance_loss_clip": 0.97918785,
      "balance_loss_mlp": 1.02311957,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 2.077867789767854,
      "language_loss": 0.76575732,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78786516,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.715397834777832
    },
    {
      "auxiliary_loss_clip": 0.01183012,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 0.97642195,
      "balance_loss_mlp": 1.01906049,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 1.7133925269283303,
      "language_loss": 0.84379387,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.8658942,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 3.6283011436462402
    },
    {
      "auxiliary_loss_clip": 0.01191267,
      "auxiliary_loss_mlp": 0.01035873,
      "balance_loss_clip": 1.0193975,
      "balance_loss_mlp": 1.02818406,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 1.6953857053555113,
      "language_loss": 0.82638359,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.84865499,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.7742793560028076
    },
    {
      "auxiliary_loss_clip": 0.01180865,
      "auxiliary_loss_mlp": 0.01022246,
      "balance_loss_clip": 1.01498723,
      "balance_loss_mlp": 1.01443839,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.067874471391789,
      "language_loss": 0.76466227,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.78669339,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.6240346431732178
    },
    {
      "auxiliary_loss_clip": 0.01187909,
      "auxiliary_loss_mlp": 0.01030229,
      "balance_loss_clip": 1.05612886,
      "balance_loss_mlp": 1.02218819,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 1.5647005490883055,
      "language_loss": 0.86459512,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.88677645,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.727639675140381
    },
    {
      "auxiliary_loss_clip": 0.01172829,
      "auxiliary_loss_mlp": 0.01030221,
      "balance_loss_clip": 0.97537398,
      "balance_loss_mlp": 1.02260375,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 1.923297317074496,
      "language_loss": 0.67267275,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69470322,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.65450382232666
    },
    {
      "auxiliary_loss_clip": 0.01187747,
      "auxiliary_loss_mlp": 0.0102996,
      "balance_loss_clip": 0.97581846,
      "balance_loss_mlp": 1.02144837,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.163981815277143,
      "language_loss": 0.78768539,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.80986243,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.7329161167144775
    },
    {
      "auxiliary_loss_clip": 0.01184851,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.01627207,
      "balance_loss_mlp": 1.02148068,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.6048446249632977,
      "language_loss": 0.84426796,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.8664127,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 2.6806411743164062
    },
    {
      "auxiliary_loss_clip": 0.0117464,
      "auxiliary_loss_mlp": 0.0102072,
      "balance_loss_clip": 0.85954243,
      "balance_loss_mlp": 1.01312661,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.1126115478268916,
      "language_loss": 0.86376095,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.88571447,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.7648098468780518
    },
    {
      "auxiliary_loss_clip": 0.01190677,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 0.98384887,
      "balance_loss_mlp": 1.02355742,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.7722126153426652,
      "language_loss": 0.7628153,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78504062,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 2.8608100414276123
    },
    {
      "auxiliary_loss_clip": 0.01183015,
      "auxiliary_loss_mlp": 0.01023909,
      "balance_loss_clip": 1.01519775,
      "balance_loss_mlp": 1.01616693,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 1.8244626799189938,
      "language_loss": 0.8921442,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91421342,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.6132988929748535
    },
    {
      "auxiliary_loss_clip": 0.01193055,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.01854205,
      "balance_loss_mlp": 1.01931798,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 1.821493482454806,
      "language_loss": 0.80252576,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.82472456,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.672626495361328
    },
    {
      "auxiliary_loss_clip": 0.01173945,
      "auxiliary_loss_mlp": 0.01024388,
      "balance_loss_clip": 0.9763031,
      "balance_loss_mlp": 1.01688647,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 2.3740749040728995,
      "language_loss": 0.70934606,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73132938,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 3.006066083908081
    },
    {
      "auxiliary_loss_clip": 0.01184375,
      "auxiliary_loss_mlp": 0.01028531,
      "balance_loss_clip": 1.01403189,
      "balance_loss_mlp": 1.02081847,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 1.5852340974778385,
      "language_loss": 0.7654615,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.78759056,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.679288625717163
    },
    {
      "auxiliary_loss_clip": 0.01176835,
      "auxiliary_loss_mlp": 0.01029027,
      "balance_loss_clip": 0.9383328,
      "balance_loss_mlp": 1.02155304,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 1.9408948748850223,
      "language_loss": 0.82079422,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.84285289,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 2.882251024246216
    },
    {
      "auxiliary_loss_clip": 0.01179125,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 0.94079316,
      "balance_loss_mlp": 1.02321386,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 1.836150722998474,
      "language_loss": 0.73208022,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.7541827,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.7167747020721436
    },
    {
      "auxiliary_loss_clip": 0.01184541,
      "auxiliary_loss_mlp": 0.01033558,
      "balance_loss_clip": 0.98233682,
      "balance_loss_mlp": 1.02638793,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 1.8671328368092395,
      "language_loss": 0.8154484,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.83762938,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 2.6630687713623047
    },
    {
      "auxiliary_loss_clip": 0.01183378,
      "auxiliary_loss_mlp": 0.01032793,
      "balance_loss_clip": 1.01757336,
      "balance_loss_mlp": 1.02530706,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 2.450352520470571,
      "language_loss": 0.86004281,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.88220453,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.675614595413208
    },
    {
      "auxiliary_loss_clip": 0.01168844,
      "auxiliary_loss_mlp": 0.01024205,
      "balance_loss_clip": 0.9720453,
      "balance_loss_mlp": 1.01626539,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.7565173812581307,
      "language_loss": 0.73981988,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76175034,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.7373502254486084
    },
    {
      "auxiliary_loss_clip": 0.01186791,
      "auxiliary_loss_mlp": 0.01030989,
      "balance_loss_clip": 1.01635456,
      "balance_loss_mlp": 1.0226388,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 1.7039518399652107,
      "language_loss": 0.78179026,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80396807,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 3.6607956886291504
    },
    {
      "auxiliary_loss_clip": 0.01091644,
      "auxiliary_loss_mlp": 0.01004333,
      "balance_loss_clip": 0.94939899,
      "balance_loss_mlp": 1.00211608,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8131323278449963,
      "language_loss": 0.66064119,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.68160099,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 4.215333461761475
    },
    {
      "auxiliary_loss_clip": 0.01184657,
      "auxiliary_loss_mlp": 0.01123624,
      "balance_loss_clip": 0.979819,
      "balance_loss_mlp": 0.0,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.315492048399918,
      "language_loss": 0.91670603,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.93978882,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.728485345840454
    },
    {
      "auxiliary_loss_clip": 0.01099436,
      "auxiliary_loss_mlp": 0.01007662,
      "balance_loss_clip": 0.92177761,
      "balance_loss_mlp": 1.00556397,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.7823033988041413,
      "language_loss": 0.54715824,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.5682292,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 3.2144484519958496
    },
    {
      "auxiliary_loss_clip": 0.01181491,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.01419258,
      "balance_loss_mlp": 1.02151799,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 1.8987138879760401,
      "language_loss": 0.80704093,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.82914853,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.7017805576324463
    },
    {
      "auxiliary_loss_clip": 0.01156512,
      "auxiliary_loss_mlp": 0.0102717,
      "balance_loss_clip": 0.74095607,
      "balance_loss_mlp": 1.01933765,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 1.7936849302262776,
      "language_loss": 0.76166004,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78349686,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 3.0109550952911377
    },
    {
      "auxiliary_loss_clip": 0.01187095,
      "auxiliary_loss_mlp": 0.01029093,
      "balance_loss_clip": 1.0561173,
      "balance_loss_mlp": 1.02147532,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 1.9925456358788394,
      "language_loss": 0.69653398,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.71869588,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 3.69246768951416
    },
    {
      "auxiliary_loss_clip": 0.01182547,
      "auxiliary_loss_mlp": 0.01026629,
      "balance_loss_clip": 0.90441716,
      "balance_loss_mlp": 1.01886249,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 2.001745911521979,
      "language_loss": 0.6618095,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.68390125,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.8162009716033936
    },
    {
      "auxiliary_loss_clip": 0.01183775,
      "auxiliary_loss_mlp": 0.01025709,
      "balance_loss_clip": 0.97635192,
      "balance_loss_mlp": 1.01773369,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 1.9044037699459564,
      "language_loss": 0.70404196,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.7261368,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 4.168538331985474
    },
    {
      "auxiliary_loss_clip": 0.01097002,
      "auxiliary_loss_mlp": 0.01003731,
      "balance_loss_clip": 0.99714184,
      "balance_loss_mlp": 1.00165701,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7916274747372686,
      "language_loss": 0.61746228,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63846958,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.2351982593536377
    },
    {
      "auxiliary_loss_clip": 0.01180512,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 0.86430258,
      "balance_loss_mlp": 1.02105689,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 6.8371526554491195,
      "language_loss": 0.76665282,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.78875291,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.808791160583496
    },
    {
      "auxiliary_loss_clip": 0.01184119,
      "auxiliary_loss_mlp": 0.01028196,
      "balance_loss_clip": 0.98203433,
      "balance_loss_mlp": 1.02067423,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 1.8527527376280946,
      "language_loss": 0.84004498,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86216819,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.7307217121124268
    },
    {
      "auxiliary_loss_clip": 0.01175858,
      "auxiliary_loss_mlp": 0.01032856,
      "balance_loss_clip": 0.97717673,
      "balance_loss_mlp": 1.0253042,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.8163645812779594,
      "language_loss": 0.76793301,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79002017,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.8322153091430664
    },
    {
      "auxiliary_loss_clip": 0.0118373,
      "auxiliary_loss_mlp": 0.01032621,
      "balance_loss_clip": 0.97703969,
      "balance_loss_mlp": 1.02381098,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 2.301788835516328,
      "language_loss": 0.67397958,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69614309,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.782687187194824
    },
    {
      "auxiliary_loss_clip": 0.01182419,
      "auxiliary_loss_mlp": 0.01033194,
      "balance_loss_clip": 1.01540327,
      "balance_loss_mlp": 1.02521276,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 1.8814592219514306,
      "language_loss": 0.77398598,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79614216,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.7122695446014404
    },
    {
      "auxiliary_loss_clip": 0.01189215,
      "auxiliary_loss_mlp": 0.01124063,
      "balance_loss_clip": 0.97650945,
      "balance_loss_mlp": 0.0,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 1.9696928540328376,
      "language_loss": 0.77866215,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.80179489,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 2.7266156673431396
    },
    {
      "auxiliary_loss_clip": 0.01188362,
      "auxiliary_loss_mlp": 0.0102753,
      "balance_loss_clip": 0.98031008,
      "balance_loss_mlp": 1.01901221,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.1291462511928394,
      "language_loss": 0.68475842,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.70691729,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.8500425815582275
    },
    {
      "auxiliary_loss_clip": 0.01182633,
      "auxiliary_loss_mlp": 0.01032155,
      "balance_loss_clip": 0.97749293,
      "balance_loss_mlp": 1.02370274,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 2.07564995538058,
      "language_loss": 0.66569215,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68784004,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.703982353210449
    },
    {
      "auxiliary_loss_clip": 0.01177122,
      "auxiliary_loss_mlp": 0.01027331,
      "balance_loss_clip": 0.9743582,
      "balance_loss_mlp": 1.01854563,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.0302536057468683,
      "language_loss": 0.85331357,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.8753581,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 2.6693837642669678
    },
    {
      "auxiliary_loss_clip": 0.01185714,
      "auxiliary_loss_mlp": 0.01029252,
      "balance_loss_clip": 0.98191249,
      "balance_loss_mlp": 1.0216409,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 1.909435098182018,
      "language_loss": 0.85756928,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.8797189,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.7014448642730713
    },
    {
      "auxiliary_loss_clip": 0.01188943,
      "auxiliary_loss_mlp": 0.01028662,
      "balance_loss_clip": 1.0146277,
      "balance_loss_mlp": 1.02034163,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 1.9487922985733113,
      "language_loss": 0.80782104,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.82999706,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.6786723136901855
    },
    {
      "auxiliary_loss_clip": 0.01188277,
      "auxiliary_loss_mlp": 0.01028872,
      "balance_loss_clip": 1.01794791,
      "balance_loss_mlp": 1.02197862,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.5455275586273625,
      "language_loss": 0.79895985,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.82113135,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 2.7227885723114014
    },
    {
      "auxiliary_loss_clip": 0.01186436,
      "auxiliary_loss_mlp": 0.01026913,
      "balance_loss_clip": 0.97622991,
      "balance_loss_mlp": 1.01886618,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 2.4727353398497742,
      "language_loss": 0.76383686,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.78597033,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.623366355895996
    },
    {
      "auxiliary_loss_clip": 0.01186494,
      "auxiliary_loss_mlp": 0.01023721,
      "balance_loss_clip": 1.05489421,
      "balance_loss_mlp": 1.01622868,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.8481753532563245,
      "language_loss": 0.8188284,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.84093058,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 2.6960949897766113
    },
    {
      "auxiliary_loss_clip": 0.01190203,
      "auxiliary_loss_mlp": 0.01030699,
      "balance_loss_clip": 1.05644739,
      "balance_loss_mlp": 1.02278948,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 1.9846689588040598,
      "language_loss": 0.70418847,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.72639751,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 2.617588758468628
    },
    {
      "auxiliary_loss_clip": 0.01179884,
      "auxiliary_loss_mlp": 0.01030659,
      "balance_loss_clip": 0.97684169,
      "balance_loss_mlp": 1.02299356,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 1.929115703112892,
      "language_loss": 0.76288879,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78499413,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.763009548187256
    },
    {
      "auxiliary_loss_clip": 0.01185236,
      "auxiliary_loss_mlp": 0.01027131,
      "balance_loss_clip": 0.97714579,
      "balance_loss_mlp": 1.01911998,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 1.7033633730232445,
      "language_loss": 0.70610404,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.72822773,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 3.7364094257354736
    },
    {
      "auxiliary_loss_clip": 0.01093513,
      "auxiliary_loss_mlp": 0.01012486,
      "balance_loss_clip": 0.98742586,
      "balance_loss_mlp": 1.01025665,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6824461424391219,
      "language_loss": 0.57864928,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59970927,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 4.219793081283569
    },
    {
      "auxiliary_loss_clip": 0.01188774,
      "auxiliary_loss_mlp": 0.01030692,
      "balance_loss_clip": 0.94055843,
      "balance_loss_mlp": 1.02240062,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 1.6233755923837085,
      "language_loss": 0.75344193,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77563655,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.727275848388672
    },
    {
      "auxiliary_loss_clip": 0.01185631,
      "auxiliary_loss_mlp": 0.01025651,
      "balance_loss_clip": 0.90211415,
      "balance_loss_mlp": 1.01759815,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 1.790856091131052,
      "language_loss": 0.83502567,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.85713851,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 2.749117851257324
    },
    {
      "auxiliary_loss_clip": 0.01183703,
      "auxiliary_loss_mlp": 0.01028037,
      "balance_loss_clip": 0.97581577,
      "balance_loss_mlp": 1.01979983,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 1.9212248487076617,
      "language_loss": 0.77923274,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80135012,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.6982810497283936
    },
    {
      "auxiliary_loss_clip": 0.01175047,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.01209819,
      "balance_loss_mlp": 1.02093542,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 1.875651311026208,
      "language_loss": 0.79956138,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82160574,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 3.685328960418701
    },
    {
      "auxiliary_loss_clip": 0.0109293,
      "auxiliary_loss_mlp": 0.0100375,
      "balance_loss_clip": 0.9498629,
      "balance_loss_mlp": 1.00149655,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 1.7239493684657128,
      "language_loss": 0.57217205,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59313881,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.3356106281280518
    },
    {
      "auxiliary_loss_clip": 0.01190168,
      "auxiliary_loss_mlp": 0.01027742,
      "balance_loss_clip": 0.97973335,
      "balance_loss_mlp": 1.01999307,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 1.6757215844193663,
      "language_loss": 0.69249773,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71467686,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.8164238929748535
    },
    {
      "auxiliary_loss_clip": 0.01170689,
      "auxiliary_loss_mlp": 0.01024093,
      "balance_loss_clip": 0.77932239,
      "balance_loss_mlp": 1.01595652,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 1.691955847218284,
      "language_loss": 0.72648859,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.74843645,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 3.916665554046631
    },
    {
      "auxiliary_loss_clip": 0.01188824,
      "auxiliary_loss_mlp": 0.01030708,
      "balance_loss_clip": 1.05626559,
      "balance_loss_mlp": 1.02255976,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 1.7128099080459087,
      "language_loss": 0.78149974,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.80369508,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.9085946083068848
    },
    {
      "auxiliary_loss_clip": 0.0118767,
      "auxiliary_loss_mlp": 0.0103009,
      "balance_loss_clip": 0.93942446,
      "balance_loss_mlp": 1.02165008,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 1.8905714745855873,
      "language_loss": 0.78110933,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80328691,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 288.1878650188446
    },
    {
      "auxiliary_loss_clip": 0.01116743,
      "auxiliary_loss_mlp": 0.01011984,
      "balance_loss_clip": 0.8513813,
      "balance_loss_mlp": 1.00976682,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.707736207170792,
      "language_loss": 0.60265124,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.6239385,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.408086061477661
    },
    {
      "auxiliary_loss_clip": 0.01089104,
      "auxiliary_loss_mlp": 0.01000762,
      "balance_loss_clip": 1.02377868,
      "balance_loss_mlp": 0.99867535,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.9698842049116833,
      "language_loss": 0.57988375,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.60078239,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 3.0478451251983643
    },
    {
      "auxiliary_loss_clip": 0.01189584,
      "auxiliary_loss_mlp": 0.01031394,
      "balance_loss_clip": 1.05756772,
      "balance_loss_mlp": 1.02359223,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.4351676395374597,
      "language_loss": 0.7585628,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.78077257,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.73270583152771
    },
    {
      "auxiliary_loss_clip": 0.01090403,
      "auxiliary_loss_mlp": 0.01002565,
      "balance_loss_clip": 0.98504996,
      "balance_loss_mlp": 1.00035942,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.7735801553117854,
      "language_loss": 0.61848772,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63941729,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.2787435054779053
    },
    {
      "auxiliary_loss_clip": 0.0118787,
      "auxiliary_loss_mlp": 0.01032519,
      "balance_loss_clip": 1.01862514,
      "balance_loss_mlp": 1.02441239,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.1748364974884904,
      "language_loss": 0.79453313,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.816737,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 2.7070703506469727
    },
    {
      "auxiliary_loss_clip": 0.01184501,
      "auxiliary_loss_mlp": 0.01124821,
      "balance_loss_clip": 0.94198716,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 3.384088974223337,
      "language_loss": 0.78830492,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.81139809,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.7297542095184326
    },
    {
      "auxiliary_loss_clip": 0.01188353,
      "auxiliary_loss_mlp": 0.01028135,
      "balance_loss_clip": 0.97621316,
      "balance_loss_mlp": 1.02036285,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 2.6412346413650845,
      "language_loss": 0.77658123,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.79874611,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.703052043914795
    },
    {
      "auxiliary_loss_clip": 0.01177775,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 0.93955976,
      "balance_loss_mlp": 1.01957917,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 2.8030552718505675,
      "language_loss": 0.74720204,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.76925159,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 2.8079516887664795
    },
    {
      "auxiliary_loss_clip": 0.01188953,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.05529606,
      "balance_loss_mlp": 1.02217567,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 1.8981492556271202,
      "language_loss": 0.7819891,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.80417573,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.6754074096679688
    },
    {
      "auxiliary_loss_clip": 0.01184367,
      "auxiliary_loss_mlp": 0.01025041,
      "balance_loss_clip": 0.9392581,
      "balance_loss_mlp": 1.01700604,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 5.271192221627894,
      "language_loss": 0.72721952,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.74931359,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 2.8276889324188232
    },
    {
      "auxiliary_loss_clip": 0.01172517,
      "auxiliary_loss_mlp": 0.01028918,
      "balance_loss_clip": 0.93543398,
      "balance_loss_mlp": 1.02086544,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 3.195867101334779,
      "language_loss": 0.79591048,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.81792486,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.8384876251220703
    },
    {
      "auxiliary_loss_clip": 0.01172118,
      "auxiliary_loss_mlp": 0.01026929,
      "balance_loss_clip": 0.9719609,
      "balance_loss_mlp": 1.01884711,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.7692074349514215,
      "language_loss": 0.76285619,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.78484666,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.7291858196258545
    },
    {
      "auxiliary_loss_clip": 0.01185667,
      "auxiliary_loss_mlp": 0.0102939,
      "balance_loss_clip": 1.0517323,
      "balance_loss_mlp": 1.02126038,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 1.814025228792301,
      "language_loss": 0.91923463,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94138515,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.7078304290771484
    },
    {
      "auxiliary_loss_clip": 0.0109298,
      "auxiliary_loss_mlp": 0.01004942,
      "balance_loss_clip": 0.90920377,
      "balance_loss_mlp": 1.00282049,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8149073589752691,
      "language_loss": 0.56532431,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58630359,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 3.418544054031372
    },
    {
      "auxiliary_loss_clip": 0.01185542,
      "auxiliary_loss_mlp": 0.01022228,
      "balance_loss_clip": 1.01695311,
      "balance_loss_mlp": 1.01406252,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 1.8164118380719938,
      "language_loss": 0.81156385,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.83364159,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.684689998626709
    },
    {
      "auxiliary_loss_clip": 0.01192627,
      "auxiliary_loss_mlp": 0.01025521,
      "balance_loss_clip": 0.97907972,
      "balance_loss_mlp": 1.0178504,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 1.5590113589555634,
      "language_loss": 0.82616508,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84834659,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 5.306323766708374
    },
    {
      "auxiliary_loss_clip": 0.0118668,
      "auxiliary_loss_mlp": 0.0102389,
      "balance_loss_clip": 1.01754165,
      "balance_loss_mlp": 1.0157783,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 1.8049124279631181,
      "language_loss": 0.70292819,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.72503388,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.7273337841033936
    },
    {
      "auxiliary_loss_clip": 0.01188259,
      "auxiliary_loss_mlp": 0.01028021,
      "balance_loss_clip": 1.01820302,
      "balance_loss_mlp": 1.02067173,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 1.8546280636918815,
      "language_loss": 0.88345331,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90561604,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.7084267139434814
    },
    {
      "auxiliary_loss_clip": 0.01193702,
      "auxiliary_loss_mlp": 0.0102655,
      "balance_loss_clip": 0.9816829,
      "balance_loss_mlp": 1.0176506,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 7.082390454054902,
      "language_loss": 0.79615462,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81835711,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 2.832284450531006
    },
    {
      "auxiliary_loss_clip": 0.01185758,
      "auxiliary_loss_mlp": 0.01027093,
      "balance_loss_clip": 0.97691453,
      "balance_loss_mlp": 1.01899874,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 1.8352706188691845,
      "language_loss": 0.81128752,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.8334161,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.781388998031616
    },
    {
      "auxiliary_loss_clip": 0.01175502,
      "auxiliary_loss_mlp": 0.01033388,
      "balance_loss_clip": 0.97587192,
      "balance_loss_mlp": 1.02582979,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 2.078964331490312,
      "language_loss": 0.63538998,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65747887,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 3.678699016571045
    },
    {
      "auxiliary_loss_clip": 0.01181939,
      "auxiliary_loss_mlp": 0.01026759,
      "balance_loss_clip": 1.01480842,
      "balance_loss_mlp": 1.01820576,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.1707272362095913,
      "language_loss": 0.74826401,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.77035099,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.7267258167266846
    },
    {
      "auxiliary_loss_clip": 0.01187052,
      "auxiliary_loss_mlp": 0.01032068,
      "balance_loss_clip": 1.05358481,
      "balance_loss_mlp": 1.02397323,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 1.9617314712952065,
      "language_loss": 0.85512829,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.87731951,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.631277322769165
    },
    {
      "auxiliary_loss_clip": 0.01180195,
      "auxiliary_loss_mlp": 0.01030461,
      "balance_loss_clip": 0.9767431,
      "balance_loss_mlp": 1.02246261,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 1.8159395913568555,
      "language_loss": 0.63810647,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66021305,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 3.6995208263397217
    },
    {
      "auxiliary_loss_clip": 0.01171648,
      "auxiliary_loss_mlp": 0.01026793,
      "balance_loss_clip": 0.89844656,
      "balance_loss_mlp": 1.01826954,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.3367382362333102,
      "language_loss": 0.91762966,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.93961406,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.8055901527404785
    },
    {
      "auxiliary_loss_clip": 0.01173613,
      "auxiliary_loss_mlp": 0.0102603,
      "balance_loss_clip": 0.93429935,
      "balance_loss_mlp": 1.01749504,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 2.0151458955266146,
      "language_loss": 0.75206506,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77406144,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.6941096782684326
    },
    {
      "auxiliary_loss_clip": 0.01177401,
      "auxiliary_loss_mlp": 0.01030792,
      "balance_loss_clip": 0.97846162,
      "balance_loss_mlp": 1.02304959,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 1.7569700270707496,
      "language_loss": 0.65445364,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67653549,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.7095909118652344
    },
    {
      "auxiliary_loss_clip": 0.01180505,
      "auxiliary_loss_mlp": 0.0103096,
      "balance_loss_clip": 0.97676343,
      "balance_loss_mlp": 1.02293134,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 2.0251950471928906,
      "language_loss": 0.67451507,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.6966297,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 2.679466724395752
    },
    {
      "auxiliary_loss_clip": 0.01175652,
      "auxiliary_loss_mlp": 0.01123823,
      "balance_loss_clip": 1.01390266,
      "balance_loss_mlp": 0.0,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 1.7641336699481711,
      "language_loss": 0.78587723,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80887198,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 2.6922221183776855
    },
    {
      "auxiliary_loss_clip": 0.011844,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.05556178,
      "balance_loss_mlp": 1.0183847,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 1.9578632888416811,
      "language_loss": 0.72214091,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74424589,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.6731162071228027
    },
    {
      "auxiliary_loss_clip": 0.01181534,
      "auxiliary_loss_mlp": 0.01025003,
      "balance_loss_clip": 1.05075383,
      "balance_loss_mlp": 1.01746297,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 1.7632947353711885,
      "language_loss": 0.85577339,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87783873,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 2.6353561878204346
    },
    {
      "auxiliary_loss_clip": 0.01178618,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.01402092,
      "balance_loss_mlp": 1.01801026,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 1.759910333752104,
      "language_loss": 0.84403813,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86607635,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.7029380798339844
    },
    {
      "auxiliary_loss_clip": 0.01186915,
      "auxiliary_loss_mlp": 0.01030234,
      "balance_loss_clip": 0.9431507,
      "balance_loss_mlp": 1.02247298,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 2.320819590265577,
      "language_loss": 0.69929719,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.72146869,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.849440097808838
    },
    {
      "auxiliary_loss_clip": 0.01170344,
      "auxiliary_loss_mlp": 0.01123772,
      "balance_loss_clip": 0.97553611,
      "balance_loss_mlp": 0.0,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 2.353318683896361,
      "language_loss": 0.73404789,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.756989,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.6570537090301514
    },
    {
      "auxiliary_loss_clip": 0.01186003,
      "auxiliary_loss_mlp": 0.01029483,
      "balance_loss_clip": 1.05411196,
      "balance_loss_mlp": 1.02135909,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.402018124852151,
      "language_loss": 0.85668725,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.87884206,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 2.651445150375366
    },
    {
      "auxiliary_loss_clip": 0.01181749,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 0.97582638,
      "balance_loss_mlp": 1.02139449,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 1.5855140250095867,
      "language_loss": 0.88169265,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.90380126,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 2.7516653537750244
    },
    {
      "auxiliary_loss_clip": 0.01184158,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 0.97488654,
      "balance_loss_mlp": 1.0235368,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.279238124474858,
      "language_loss": 0.85042691,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.87258077,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 2.721346139907837
    },
    {
      "auxiliary_loss_clip": 0.01185752,
      "auxiliary_loss_mlp": 0.01030643,
      "balance_loss_clip": 1.01633072,
      "balance_loss_mlp": 1.0224117,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.182295355999049,
      "language_loss": 0.78894842,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.8111124,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.7207090854644775
    },
    {
      "auxiliary_loss_clip": 0.01190485,
      "auxiliary_loss_mlp": 0.011242,
      "balance_loss_clip": 0.94105041,
      "balance_loss_mlp": 0.0,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 1.4616557850912124,
      "language_loss": 0.80659431,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.82974112,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 2.753560781478882
    },
    {
      "auxiliary_loss_clip": 0.01170791,
      "auxiliary_loss_mlp": 0.01028157,
      "balance_loss_clip": 0.97455049,
      "balance_loss_mlp": 1.0203371,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 1.8371630484504444,
      "language_loss": 0.75198728,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.77397674,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 2.664545774459839
    },
    {
      "auxiliary_loss_clip": 0.01187251,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.05399942,
      "balance_loss_mlp": 1.01720214,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.6730784534556373,
      "language_loss": 0.86530203,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.88743025,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.777749538421631
    },
    {
      "auxiliary_loss_clip": 0.01190706,
      "auxiliary_loss_mlp": 0.01030822,
      "balance_loss_clip": 1.01365912,
      "balance_loss_mlp": 1.02220368,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 1.803984887207016,
      "language_loss": 0.85759825,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.87981355,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 3.645355224609375
    },
    {
      "auxiliary_loss_clip": 0.01186607,
      "auxiliary_loss_mlp": 0.01026502,
      "balance_loss_clip": 1.05277419,
      "balance_loss_mlp": 1.01816308,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 1.8530926250628008,
      "language_loss": 0.73006988,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75220096,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 3.5870485305786133
    },
    {
      "auxiliary_loss_clip": 0.01190681,
      "auxiliary_loss_mlp": 0.0102593,
      "balance_loss_clip": 1.01616466,
      "balance_loss_mlp": 1.01780641,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.2845734085002736,
      "language_loss": 0.74908376,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.77124989,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.6914467811584473
    },
    {
      "auxiliary_loss_clip": 0.01181643,
      "auxiliary_loss_mlp": 0.01036914,
      "balance_loss_clip": 0.98129439,
      "balance_loss_mlp": 1.02845621,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.1617022094538534,
      "language_loss": 0.80955249,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83173811,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.6334521770477295
    },
    {
      "auxiliary_loss_clip": 0.01184797,
      "auxiliary_loss_mlp": 0.01030613,
      "balance_loss_clip": 1.05365479,
      "balance_loss_mlp": 1.0231564,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 1.8558091557026393,
      "language_loss": 0.74734795,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.76950204,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 2.616793155670166
    },
    {
      "auxiliary_loss_clip": 0.01183951,
      "auxiliary_loss_mlp": 0.01027839,
      "balance_loss_clip": 1.01576519,
      "balance_loss_mlp": 1.01943529,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 2.745190519190429,
      "language_loss": 0.80511445,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82723236,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.653526782989502
    },
    {
      "auxiliary_loss_clip": 0.011804,
      "auxiliary_loss_mlp": 0.01029496,
      "balance_loss_clip": 0.93891954,
      "balance_loss_mlp": 1.02109766,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 1.971860594673484,
      "language_loss": 0.76396805,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78606701,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 3.6608500480651855
    },
    {
      "auxiliary_loss_clip": 0.01187114,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.05491924,
      "balance_loss_mlp": 1.01559734,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 1.6914666523260038,
      "language_loss": 0.73901212,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.76111388,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.7085824012756348
    },
    {
      "auxiliary_loss_clip": 0.01177114,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 0.93725359,
      "balance_loss_mlp": 1.02069712,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 1.7041071589832308,
      "language_loss": 0.82836014,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.85041952,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 3.6300671100616455
    },
    {
      "auxiliary_loss_clip": 0.01092335,
      "auxiliary_loss_mlp": 0.01001209,
      "balance_loss_clip": 0.91213375,
      "balance_loss_mlp": 0.99906313,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7890877020649616,
      "language_loss": 0.54852223,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56945765,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.178077459335327
    },
    {
      "auxiliary_loss_clip": 0.01185183,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.01758635,
      "balance_loss_mlp": 1.02632344,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.2057168026948215,
      "language_loss": 0.75915337,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78134882,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.7398030757904053
    },
    {
      "auxiliary_loss_clip": 0.01191798,
      "auxiliary_loss_mlp": 0.01030653,
      "balance_loss_clip": 1.01866698,
      "balance_loss_mlp": 1.0222311,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 1.5685804197388038,
      "language_loss": 0.71786183,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.74008638,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.708951234817505
    },
    {
      "auxiliary_loss_clip": 0.01177746,
      "auxiliary_loss_mlp": 0.01027862,
      "balance_loss_clip": 0.97279602,
      "balance_loss_mlp": 1.02050126,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 1.5544497012353553,
      "language_loss": 0.82604265,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.8480987,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.7863216400146484
    },
    {
      "auxiliary_loss_clip": 0.01185008,
      "auxiliary_loss_mlp": 0.01037365,
      "balance_loss_clip": 0.97399235,
      "balance_loss_mlp": 1.02934813,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.800202896770222,
      "language_loss": 0.85938174,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.88160551,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.7578675746917725
    },
    {
      "auxiliary_loss_clip": 0.01169085,
      "auxiliary_loss_mlp": 0.01027332,
      "balance_loss_clip": 0.97267103,
      "balance_loss_mlp": 1.01988149,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 1.56503367038316,
      "language_loss": 0.73226249,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.75422668,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.7720346450805664
    },
    {
      "auxiliary_loss_clip": 0.0118151,
      "auxiliary_loss_mlp": 0.01021405,
      "balance_loss_clip": 1.01505411,
      "balance_loss_mlp": 1.01398969,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 2.051782743736697,
      "language_loss": 0.85268736,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87471652,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 2.608196973800659
    },
    {
      "auxiliary_loss_clip": 0.01180185,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 0.90014791,
      "balance_loss_mlp": 1.01984286,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.4723096379109974,
      "language_loss": 0.80697256,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.82905459,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.8514811992645264
    },
    {
      "auxiliary_loss_clip": 0.01174148,
      "auxiliary_loss_mlp": 0.01027682,
      "balance_loss_clip": 0.97217602,
      "balance_loss_mlp": 1.01943266,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 2.3982494499474267,
      "language_loss": 0.73969162,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.76170993,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.6862990856170654
    },
    {
      "auxiliary_loss_clip": 0.01184644,
      "auxiliary_loss_mlp": 0.01030774,
      "balance_loss_clip": 1.0527333,
      "balance_loss_mlp": 1.0229423,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 1.8439749343611571,
      "language_loss": 0.77686077,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.79901499,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.688330888748169
    },
    {
      "auxiliary_loss_clip": 0.0118295,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 1.01296246,
      "balance_loss_mlp": 1.01757908,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 1.8245372352308444,
      "language_loss": 0.76619089,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.78827608,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.6781582832336426
    },
    {
      "auxiliary_loss_clip": 0.01088695,
      "auxiliary_loss_mlp": 0.00999873,
      "balance_loss_clip": 1.02451849,
      "balance_loss_mlp": 0.99785846,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7273286775114532,
      "language_loss": 0.56736892,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58825463,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 3.117621421813965
    },
    {
      "auxiliary_loss_clip": 0.01172863,
      "auxiliary_loss_mlp": 0.01029507,
      "balance_loss_clip": 1.01102901,
      "balance_loss_mlp": 1.02156162,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.7972439887811875,
      "language_loss": 0.81693125,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.83895493,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.702841281890869
    },
    {
      "auxiliary_loss_clip": 0.011768,
      "auxiliary_loss_mlp": 0.01025686,
      "balance_loss_clip": 0.97470939,
      "balance_loss_mlp": 1.01826191,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 1.8031331682476954,
      "language_loss": 0.71912593,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.74115074,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.7359704971313477
    },
    {
      "auxiliary_loss_clip": 0.01184247,
      "auxiliary_loss_mlp": 0.01028825,
      "balance_loss_clip": 1.01425624,
      "balance_loss_mlp": 1.02132726,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 2.15384942071174,
      "language_loss": 0.75905383,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78118455,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 2.7211570739746094
    },
    {
      "auxiliary_loss_clip": 0.01189751,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.02040577,
      "balance_loss_mlp": 1.01934004,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.434001929791834,
      "language_loss": 0.83053601,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.85271382,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.6463730335235596
    },
    {
      "auxiliary_loss_clip": 0.01183116,
      "auxiliary_loss_mlp": 0.0112401,
      "balance_loss_clip": 0.93914258,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.6868832115648362,
      "language_loss": 0.77958047,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.8026517,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 2.74125599861145
    },
    {
      "auxiliary_loss_clip": 0.01180571,
      "auxiliary_loss_mlp": 0.01123811,
      "balance_loss_clip": 1.01765966,
      "balance_loss_mlp": 0.0,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 1.8848083245666567,
      "language_loss": 0.86366594,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.88670981,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 4.6604063510894775
    },
    {
      "auxiliary_loss_clip": 0.01173759,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 1.0128032,
      "balance_loss_mlp": 1.02173591,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.400074865155392,
      "language_loss": 0.75607824,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.77811718,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.6175432205200195
    },
    {
      "auxiliary_loss_clip": 0.01171132,
      "auxiliary_loss_mlp": 0.01024752,
      "balance_loss_clip": 0.86282533,
      "balance_loss_mlp": 1.01659822,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.5387914619840795,
      "language_loss": 0.82608461,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.84804344,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.908864736557007
    },
    {
      "auxiliary_loss_clip": 0.01099244,
      "auxiliary_loss_mlp": 0.00999253,
      "balance_loss_clip": 0.83513021,
      "balance_loss_mlp": 0.99703592,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7771672907880406,
      "language_loss": 0.605299,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.626284,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.564884662628174
    },
    {
      "auxiliary_loss_clip": 0.01179304,
      "auxiliary_loss_mlp": 0.01031684,
      "balance_loss_clip": 0.93706888,
      "balance_loss_mlp": 1.02379835,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 1.56793700425882,
      "language_loss": 0.76921678,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79132658,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 3.059183120727539
    },
    {
      "auxiliary_loss_clip": 0.01184562,
      "auxiliary_loss_mlp": 0.01028992,
      "balance_loss_clip": 1.01371574,
      "balance_loss_mlp": 1.02045107,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 1.5717750098360384,
      "language_loss": 0.77887732,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.80101287,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 2.7273755073547363
    },
    {
      "auxiliary_loss_clip": 0.01173765,
      "auxiliary_loss_mlp": 0.0102418,
      "balance_loss_clip": 0.89627004,
      "balance_loss_mlp": 1.01605558,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 4.948926218135286,
      "language_loss": 0.70060873,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.72258824,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 3.6305599212646484
    },
    {
      "auxiliary_loss_clip": 0.01182379,
      "auxiliary_loss_mlp": 0.01031398,
      "balance_loss_clip": 0.89765102,
      "balance_loss_mlp": 1.02279675,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.039472003565162,
      "language_loss": 0.78432244,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80646014,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.768949270248413
    },
    {
      "auxiliary_loss_clip": 0.01183168,
      "auxiliary_loss_mlp": 0.01026137,
      "balance_loss_clip": 1.05126095,
      "balance_loss_mlp": 1.01815557,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 3.1441503687670087,
      "language_loss": 0.83138752,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85348058,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 3.9625837802886963
    },
    {
      "auxiliary_loss_clip": 0.01183378,
      "auxiliary_loss_mlp": 0.01027889,
      "balance_loss_clip": 0.97455478,
      "balance_loss_mlp": 1.01998544,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 3.2665337005419897,
      "language_loss": 0.69420779,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.7163204,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.726911783218384
    },
    {
      "auxiliary_loss_clip": 0.0118221,
      "auxiliary_loss_mlp": 0.01032755,
      "balance_loss_clip": 1.01515818,
      "balance_loss_mlp": 1.02495253,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.0689021114228163,
      "language_loss": 0.70168626,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72383589,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 2.703693389892578
    },
    {
      "auxiliary_loss_clip": 0.01183757,
      "auxiliary_loss_mlp": 0.01029915,
      "balance_loss_clip": 0.97541451,
      "balance_loss_mlp": 1.02171636,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 1.9249181719942439,
      "language_loss": 0.77134931,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.79348606,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.657454490661621
    },
    {
      "auxiliary_loss_clip": 0.01081162,
      "auxiliary_loss_mlp": 0.01003611,
      "balance_loss_clip": 0.94200146,
      "balance_loss_mlp": 1.00154901,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8425180014080119,
      "language_loss": 0.6517182,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67256594,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.3524019718170166
    },
    {
      "auxiliary_loss_clip": 0.01185285,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.01450849,
      "balance_loss_mlp": 1.01844096,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 1.711521549076087,
      "language_loss": 0.7565425,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.77866083,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.694714307785034
    },
    {
      "auxiliary_loss_clip": 0.01174638,
      "auxiliary_loss_mlp": 0.01034099,
      "balance_loss_clip": 0.89780807,
      "balance_loss_mlp": 1.02621961,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.1160929840079774,
      "language_loss": 0.89618117,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.9182685,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.744387149810791
    },
    {
      "auxiliary_loss_clip": 0.01176705,
      "auxiliary_loss_mlp": 0.0102939,
      "balance_loss_clip": 0.89765012,
      "balance_loss_mlp": 1.0215162,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 1.7672457987221066,
      "language_loss": 0.81047428,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83253521,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.8496129512786865
    },
    {
      "auxiliary_loss_clip": 0.01182672,
      "auxiliary_loss_mlp": 0.01033795,
      "balance_loss_clip": 1.053514,
      "balance_loss_mlp": 1.02615976,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 2.2500728537054706,
      "language_loss": 0.69225538,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71441996,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 2.7674684524536133
    },
    {
      "auxiliary_loss_clip": 0.01170823,
      "auxiliary_loss_mlp": 0.01028571,
      "balance_loss_clip": 0.866009,
      "balance_loss_mlp": 1.02098966,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 1.8252423099620834,
      "language_loss": 0.79522693,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.81722093,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 2.7828848361968994
    },
    {
      "auxiliary_loss_clip": 0.01176351,
      "auxiliary_loss_mlp": 0.01030469,
      "balance_loss_clip": 0.93847775,
      "balance_loss_mlp": 1.02166569,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.659966179302985,
      "language_loss": 0.68947512,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71154332,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.7649991512298584
    },
    {
      "auxiliary_loss_clip": 0.01186566,
      "auxiliary_loss_mlp": 0.01026807,
      "balance_loss_clip": 0.93746787,
      "balance_loss_mlp": 1.01852155,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 1.5134863658217417,
      "language_loss": 0.69978029,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.72191405,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.7923877239227295
    },
    {
      "auxiliary_loss_clip": 0.01184344,
      "auxiliary_loss_mlp": 0.01023311,
      "balance_loss_clip": 1.0130322,
      "balance_loss_mlp": 1.01537776,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.7127429586293936,
      "language_loss": 0.83368063,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.85575724,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 2.735280752182007
    },
    {
      "auxiliary_loss_clip": 0.01179609,
      "auxiliary_loss_mlp": 0.01023862,
      "balance_loss_clip": 1.05272317,
      "balance_loss_mlp": 1.01698661,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 2.126247044513198,
      "language_loss": 0.87654018,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.89857483,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 2.6166374683380127
    },
    {
      "auxiliary_loss_clip": 0.01171106,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 0.93633473,
      "balance_loss_mlp": 1.01653385,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 1.8668011665501323,
      "language_loss": 0.91175067,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93370837,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.743934392929077
    },
    {
      "auxiliary_loss_clip": 0.01178505,
      "auxiliary_loss_mlp": 0.01029885,
      "balance_loss_clip": 0.93590438,
      "balance_loss_mlp": 1.02168965,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.2502130815436017,
      "language_loss": 0.66578841,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.68787241,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.6952548027038574
    },
    {
      "auxiliary_loss_clip": 0.01184142,
      "auxiliary_loss_mlp": 0.01031522,
      "balance_loss_clip": 1.05470657,
      "balance_loss_mlp": 1.02383268,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.5056104674050237,
      "language_loss": 0.78127778,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80343437,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 2.6423511505126953
    },
    {
      "auxiliary_loss_clip": 0.01183933,
      "auxiliary_loss_mlp": 0.0103126,
      "balance_loss_clip": 1.05443144,
      "balance_loss_mlp": 1.02298427,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 1.7639245942000745,
      "language_loss": 0.69821024,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.72036219,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 2.6186821460723877
    },
    {
      "auxiliary_loss_clip": 0.0117851,
      "auxiliary_loss_mlp": 0.01028921,
      "balance_loss_clip": 0.97250867,
      "balance_loss_mlp": 1.02031434,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 2.0174833704749853,
      "language_loss": 0.66787755,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.6899519,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 3.7832915782928467
    },
    {
      "auxiliary_loss_clip": 0.01182254,
      "auxiliary_loss_mlp": 0.01123507,
      "balance_loss_clip": 0.97662181,
      "balance_loss_mlp": 0.0,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 2.5464336783145636,
      "language_loss": 0.64350247,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.66656011,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.747412919998169
    },
    {
      "auxiliary_loss_clip": 0.01184624,
      "auxiliary_loss_mlp": 0.0103302,
      "balance_loss_clip": 0.93436849,
      "balance_loss_mlp": 1.02467561,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 2.1486091239690865,
      "language_loss": 0.87857699,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.90075344,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.7301039695739746
    },
    {
      "auxiliary_loss_clip": 0.01178038,
      "auxiliary_loss_mlp": 0.01032436,
      "balance_loss_clip": 0.97532666,
      "balance_loss_mlp": 1.02364397,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 1.8011583307791093,
      "language_loss": 0.78466201,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.80676675,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.763554573059082
    },
    {
      "auxiliary_loss_clip": 0.01175769,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 0.97170925,
      "balance_loss_mlp": 1.01954472,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 2.552514996550449,
      "language_loss": 0.85193551,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.87397313,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.679291248321533
    },
    {
      "auxiliary_loss_clip": 0.01183711,
      "auxiliary_loss_mlp": 0.01028272,
      "balance_loss_clip": 0.94101179,
      "balance_loss_mlp": 1.02026689,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 2.0014833917837245,
      "language_loss": 0.85197973,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.87409961,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 3.728872776031494
    },
    {
      "auxiliary_loss_clip": 0.01173445,
      "auxiliary_loss_mlp": 0.0102924,
      "balance_loss_clip": 0.97508669,
      "balance_loss_mlp": 1.02118754,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 1.7413340400418342,
      "language_loss": 0.82276243,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84478927,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.7063980102539062
    },
    {
      "auxiliary_loss_clip": 0.01188904,
      "auxiliary_loss_mlp": 0.01027751,
      "balance_loss_clip": 0.97551596,
      "balance_loss_mlp": 1.01958513,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 2.1284224597187427,
      "language_loss": 0.79926133,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.82142788,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.7632999420166016
    },
    {
      "auxiliary_loss_clip": 0.01183506,
      "auxiliary_loss_mlp": 0.01035699,
      "balance_loss_clip": 1.01531184,
      "balance_loss_mlp": 1.02744424,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.1218474333259048,
      "language_loss": 0.79339981,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81559181,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 3.5941896438598633
    },
    {
      "auxiliary_loss_clip": 0.0117184,
      "auxiliary_loss_mlp": 0.01024654,
      "balance_loss_clip": 0.97681487,
      "balance_loss_mlp": 1.01698923,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 2.176032021267838,
      "language_loss": 0.69059181,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.71255672,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.718360662460327
    },
    {
      "auxiliary_loss_clip": 0.01189023,
      "auxiliary_loss_mlp": 0.0102881,
      "balance_loss_clip": 1.01716375,
      "balance_loss_mlp": 1.02106082,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.375063941035949,
      "language_loss": 0.76875591,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.79093421,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.696289539337158
    },
    {
      "auxiliary_loss_clip": 0.01165463,
      "auxiliary_loss_mlp": 0.01030815,
      "balance_loss_clip": 0.93355691,
      "balance_loss_mlp": 1.02310181,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 2.2587815684218016,
      "language_loss": 0.59142768,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.61339045,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.6588878631591797
    },
    {
      "auxiliary_loss_clip": 0.01187719,
      "auxiliary_loss_mlp": 0.01025308,
      "balance_loss_clip": 1.0144068,
      "balance_loss_mlp": 1.01712465,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 1.6943363256158173,
      "language_loss": 0.73182106,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.75395143,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.7496204376220703
    },
    {
      "auxiliary_loss_clip": 0.01171491,
      "auxiliary_loss_mlp": 0.0102993,
      "balance_loss_clip": 0.93829441,
      "balance_loss_mlp": 1.02126312,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.096196054495243,
      "language_loss": 0.74847937,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77049351,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.697836399078369
    },
    {
      "auxiliary_loss_clip": 0.01180786,
      "auxiliary_loss_mlp": 0.01024846,
      "balance_loss_clip": 1.01170301,
      "balance_loss_mlp": 1.01750863,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 1.6878006559054126,
      "language_loss": 0.85231209,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87436843,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.7630937099456787
    },
    {
      "auxiliary_loss_clip": 0.01182003,
      "auxiliary_loss_mlp": 0.01035549,
      "balance_loss_clip": 0.93696076,
      "balance_loss_mlp": 1.02752614,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 2.108063687135923,
      "language_loss": 0.74563026,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.76780576,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 2.86586856842041
    },
    {
      "auxiliary_loss_clip": 0.01179838,
      "auxiliary_loss_mlp": 0.01030574,
      "balance_loss_clip": 0.93902886,
      "balance_loss_mlp": 1.02254558,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 2.5778392650318454,
      "language_loss": 0.82574129,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84784532,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 2.7733356952667236
    },
    {
      "auxiliary_loss_clip": 0.01185929,
      "auxiliary_loss_mlp": 0.01028871,
      "balance_loss_clip": 0.98007399,
      "balance_loss_mlp": 1.02024615,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 2.250102537687514,
      "language_loss": 0.75266922,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.77481723,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 2.763821840286255
    },
    {
      "auxiliary_loss_clip": 0.0118678,
      "auxiliary_loss_mlp": 0.01024268,
      "balance_loss_clip": 1.05546165,
      "balance_loss_mlp": 1.01641822,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 2.3090699830268733,
      "language_loss": 0.83651906,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.85862958,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.589370012283325
    },
    {
      "auxiliary_loss_clip": 0.01177475,
      "auxiliary_loss_mlp": 0.01022075,
      "balance_loss_clip": 0.97427988,
      "balance_loss_mlp": 1.01447284,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.4439809363161333,
      "language_loss": 0.83893186,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86092734,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.7728116512298584
    },
    {
      "auxiliary_loss_clip": 0.01176329,
      "auxiliary_loss_mlp": 0.01027214,
      "balance_loss_clip": 0.89890981,
      "balance_loss_mlp": 1.0190897,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 2.3085644408826727,
      "language_loss": 0.72567821,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.74771369,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 2.7296056747436523
    },
    {
      "auxiliary_loss_clip": 0.01186874,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.05478287,
      "balance_loss_mlp": 1.02150548,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 2.5226314209194873,
      "language_loss": 0.74912602,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.771294,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 2.800088405609131
    },
    {
      "auxiliary_loss_clip": 0.01184351,
      "auxiliary_loss_mlp": 0.01038881,
      "balance_loss_clip": 1.05254197,
      "balance_loss_mlp": 1.03078127,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 1.5549155968185346,
      "language_loss": 0.71732032,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.73955262,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 2.6331863403320312
    },
    {
      "auxiliary_loss_clip": 0.01186978,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.01397824,
      "balance_loss_mlp": 1.01890826,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 2.051412427661836,
      "language_loss": 0.76035964,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78249824,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.704639196395874
    },
    {
      "auxiliary_loss_clip": 0.01176729,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 0.94105232,
      "balance_loss_mlp": 1.01783299,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 1.6467592433166236,
      "language_loss": 0.81631196,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83833086,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 2.7701478004455566
    },
    {
      "auxiliary_loss_clip": 0.01187367,
      "auxiliary_loss_mlp": 0.0103291,
      "balance_loss_clip": 1.05572724,
      "balance_loss_mlp": 1.02579331,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 2.1910146096019383,
      "language_loss": 0.75094092,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77314377,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.6890008449554443
    },
    {
      "auxiliary_loss_clip": 0.01185196,
      "auxiliary_loss_mlp": 0.01030216,
      "balance_loss_clip": 1.0155561,
      "balance_loss_mlp": 1.02180624,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.5950852612207989,
      "language_loss": 0.80150449,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82365865,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 3.58577561378479
    },
    {
      "auxiliary_loss_clip": 0.01183992,
      "auxiliary_loss_mlp": 0.01025008,
      "balance_loss_clip": 0.97682059,
      "balance_loss_mlp": 1.0174557,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.7914680161359797,
      "language_loss": 0.72970951,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75179946,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 3.721907138824463
    },
    {
      "auxiliary_loss_clip": 0.01184097,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.05385292,
      "balance_loss_mlp": 1.01759028,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 1.7175818558574123,
      "language_loss": 0.70972228,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73181731,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.7651565074920654
    },
    {
      "auxiliary_loss_clip": 0.01181803,
      "auxiliary_loss_mlp": 0.01030191,
      "balance_loss_clip": 1.01354039,
      "balance_loss_mlp": 1.0224905,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 2.58411783519183,
      "language_loss": 0.75685728,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.77897722,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.681755542755127
    },
    {
      "auxiliary_loss_clip": 0.01187435,
      "auxiliary_loss_mlp": 0.01029688,
      "balance_loss_clip": 0.97945929,
      "balance_loss_mlp": 1.02217245,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.7585968894645216,
      "language_loss": 0.86309844,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.88526976,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.7314984798431396
    },
    {
      "auxiliary_loss_clip": 0.01180812,
      "auxiliary_loss_mlp": 0.01029253,
      "balance_loss_clip": 0.97783935,
      "balance_loss_mlp": 1.02026439,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 2.0713743474238084,
      "language_loss": 0.71176648,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.73386717,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.7406585216522217
    },
    {
      "auxiliary_loss_clip": 0.0118503,
      "auxiliary_loss_mlp": 0.01033866,
      "balance_loss_clip": 0.97775888,
      "balance_loss_mlp": 1.02607524,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 1.8542563290761596,
      "language_loss": 0.77398521,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79617417,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 3.548403024673462
    },
    {
      "auxiliary_loss_clip": 0.01175482,
      "auxiliary_loss_mlp": 0.01029,
      "balance_loss_clip": 0.93517041,
      "balance_loss_mlp": 1.02118599,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.201345169196974,
      "language_loss": 0.83041418,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.85245895,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.7635326385498047
    },
    {
      "auxiliary_loss_clip": 0.01186436,
      "auxiliary_loss_mlp": 0.01027628,
      "balance_loss_clip": 1.0575608,
      "balance_loss_mlp": 1.02001023,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 1.9553808626044562,
      "language_loss": 0.65124154,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.67338216,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 3.647773265838623
    },
    {
      "auxiliary_loss_clip": 0.01182465,
      "auxiliary_loss_mlp": 0.01030586,
      "balance_loss_clip": 1.01517034,
      "balance_loss_mlp": 1.02323127,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 1.6439877765467155,
      "language_loss": 0.67805564,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.70018613,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.693096399307251
    },
    {
      "auxiliary_loss_clip": 0.01171316,
      "auxiliary_loss_mlp": 0.01024164,
      "balance_loss_clip": 0.89795536,
      "balance_loss_mlp": 1.01587939,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 2.2448103470622973,
      "language_loss": 0.76879984,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.79075468,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.7907612323760986
    },
    {
      "auxiliary_loss_clip": 0.0116891,
      "auxiliary_loss_mlp": 0.01124271,
      "balance_loss_clip": 0.93550289,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.05163983420307,
      "language_loss": 0.74756902,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.77050084,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.715069055557251
    },
    {
      "auxiliary_loss_clip": 0.01178129,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 0.8993057,
      "balance_loss_mlp": 1.02182913,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 2.6343050376991646,
      "language_loss": 0.73852915,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.760611,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.7493739128112793
    },
    {
      "auxiliary_loss_clip": 0.01180706,
      "auxiliary_loss_mlp": 0.01027554,
      "balance_loss_clip": 0.97599787,
      "balance_loss_mlp": 1.01969171,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.9053996479775888,
      "language_loss": 0.87255692,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89463949,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 2.7053873538970947
    },
    {
      "auxiliary_loss_clip": 0.01155389,
      "auxiliary_loss_mlp": 0.01034525,
      "balance_loss_clip": 0.85618281,
      "balance_loss_mlp": 1.02649069,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 1.5955692931938328,
      "language_loss": 0.78970343,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.81160259,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.8523082733154297
    },
    {
      "auxiliary_loss_clip": 0.01182779,
      "auxiliary_loss_mlp": 0.01027026,
      "balance_loss_clip": 1.01431274,
      "balance_loss_mlp": 1.01902676,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 2.6365375437842444,
      "language_loss": 0.7869004,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.80899835,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.6426756381988525
    },
    {
      "auxiliary_loss_clip": 0.01185937,
      "auxiliary_loss_mlp": 0.01033278,
      "balance_loss_clip": 1.01518512,
      "balance_loss_mlp": 1.02528536,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 2.243396896160623,
      "language_loss": 0.70947182,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.73166394,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.790417432785034
    },
    {
      "auxiliary_loss_clip": 0.0117618,
      "auxiliary_loss_mlp": 0.0103238,
      "balance_loss_clip": 1.01417327,
      "balance_loss_mlp": 1.02432156,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 1.912735639636027,
      "language_loss": 0.82863563,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.85072124,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 2.7321414947509766
    },
    {
      "auxiliary_loss_clip": 0.01175941,
      "auxiliary_loss_mlp": 0.01034511,
      "balance_loss_clip": 0.97485095,
      "balance_loss_mlp": 1.02618408,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.613615760964088,
      "language_loss": 0.7528314,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77493584,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.6975929737091064
    },
    {
      "auxiliary_loss_clip": 0.01188018,
      "auxiliary_loss_mlp": 0.01123977,
      "balance_loss_clip": 1.05724335,
      "balance_loss_mlp": 0.0,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 1.5863018950070351,
      "language_loss": 0.66523474,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.68835473,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.7244601249694824
    },
    {
      "auxiliary_loss_clip": 0.01184226,
      "auxiliary_loss_mlp": 0.01029323,
      "balance_loss_clip": 1.01653469,
      "balance_loss_mlp": 1.02081704,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 1.9526978365070122,
      "language_loss": 0.68719625,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.70933175,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 2.8193888664245605
    },
    {
      "auxiliary_loss_clip": 0.01185802,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.05403996,
      "balance_loss_mlp": 1.01957989,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 2.5566045033970926,
      "language_loss": 0.6154145,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63754469,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.780707836151123
    },
    {
      "auxiliary_loss_clip": 0.011692,
      "auxiliary_loss_mlp": 0.01027354,
      "balance_loss_clip": 0.85903454,
      "balance_loss_mlp": 1.01935482,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 1.7534449073021712,
      "language_loss": 0.79231644,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81428206,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.7944750785827637
    },
    {
      "auxiliary_loss_clip": 0.0117961,
      "auxiliary_loss_mlp": 0.01031849,
      "balance_loss_clip": 1.01532507,
      "balance_loss_mlp": 1.02359939,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.6231094612164216,
      "language_loss": 0.74799752,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.7701121,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 2.716191291809082
    },
    {
      "auxiliary_loss_clip": 0.01192279,
      "auxiliary_loss_mlp": 0.01027575,
      "balance_loss_clip": 1.0187782,
      "balance_loss_mlp": 1.01973069,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 2.2180529369216426,
      "language_loss": 0.68510842,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.70730698,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 3.0702743530273438
    },
    {
      "auxiliary_loss_clip": 0.01169793,
      "auxiliary_loss_mlp": 0.01031761,
      "balance_loss_clip": 0.93572646,
      "balance_loss_mlp": 1.02334511,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 1.8898162333862962,
      "language_loss": 0.87658709,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.8986026,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.766219139099121
    },
    {
      "auxiliary_loss_clip": 0.01188211,
      "auxiliary_loss_mlp": 0.01032427,
      "balance_loss_clip": 1.0187273,
      "balance_loss_mlp": 1.02387393,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.081988231652057,
      "language_loss": 0.71309793,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.7353043,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 3.518368721008301
    },
    {
      "auxiliary_loss_clip": 0.01172146,
      "auxiliary_loss_mlp": 0.01124678,
      "balance_loss_clip": 0.89579505,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 2.30655523152882,
      "language_loss": 0.83603007,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.8589983,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 2.782423734664917
    },
    {
      "auxiliary_loss_clip": 0.01186894,
      "auxiliary_loss_mlp": 0.0102474,
      "balance_loss_clip": 1.05534959,
      "balance_loss_mlp": 1.01733434,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.2310911579788733,
      "language_loss": 0.91828656,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94040293,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.6704111099243164
    },
    {
      "auxiliary_loss_clip": 0.01185528,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 0.93924105,
      "balance_loss_mlp": 1.02029872,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 1.8586598616053127,
      "language_loss": 0.7465117,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.76864552,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.7435412406921387
    },
    {
      "auxiliary_loss_clip": 0.0118864,
      "auxiliary_loss_mlp": 0.01021841,
      "balance_loss_clip": 1.01928341,
      "balance_loss_mlp": 1.01346636,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 2.0494332870072736,
      "language_loss": 0.70864582,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73075062,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.7516915798187256
    },
    {
      "auxiliary_loss_clip": 0.01189323,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.01610899,
      "balance_loss_mlp": 1.01866484,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.498268287327362,
      "language_loss": 0.76386726,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.78602779,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 3.6576428413391113
    },
    {
      "auxiliary_loss_clip": 0.01190357,
      "auxiliary_loss_mlp": 0.01033582,
      "balance_loss_clip": 1.05516124,
      "balance_loss_mlp": 1.0249331,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.374519901444704,
      "language_loss": 0.8031137,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.82535303,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.6042940616607666
    },
    {
      "auxiliary_loss_clip": 0.01191481,
      "auxiliary_loss_mlp": 0.01029664,
      "balance_loss_clip": 1.02264571,
      "balance_loss_mlp": 1.0216893,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.6532132809500395,
      "language_loss": 0.82392371,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.84613508,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.6588799953460693
    },
    {
      "auxiliary_loss_clip": 0.01181036,
      "auxiliary_loss_mlp": 0.01026044,
      "balance_loss_clip": 0.93963552,
      "balance_loss_mlp": 1.01856947,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 1.9473933958747376,
      "language_loss": 0.76702547,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.78909624,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 3.564392328262329
    },
    {
      "auxiliary_loss_clip": 0.01185967,
      "auxiliary_loss_mlp": 0.01025585,
      "balance_loss_clip": 1.05544412,
      "balance_loss_mlp": 1.0176878,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.4680304474591028,
      "language_loss": 0.72634679,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.74846232,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.781047821044922
    },
    {
      "auxiliary_loss_clip": 0.01190185,
      "auxiliary_loss_mlp": 0.01030046,
      "balance_loss_clip": 1.05505335,
      "balance_loss_mlp": 1.02177906,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 2.0737394428801568,
      "language_loss": 0.8160212,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.83822346,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.661757707595825
    },
    {
      "auxiliary_loss_clip": 0.01193782,
      "auxiliary_loss_mlp": 0.01036977,
      "balance_loss_clip": 1.05909431,
      "balance_loss_mlp": 1.02855539,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 1.7101408291939928,
      "language_loss": 0.83942688,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86173451,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.629390239715576
    },
    {
      "auxiliary_loss_clip": 0.01171367,
      "auxiliary_loss_mlp": 0.01021986,
      "balance_loss_clip": 1.01191545,
      "balance_loss_mlp": 1.01448512,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 1.7876619961697222,
      "language_loss": 0.82806325,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.84999681,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.693220376968384
    },
    {
      "auxiliary_loss_clip": 0.01177923,
      "auxiliary_loss_mlp": 0.01027063,
      "balance_loss_clip": 0.97557545,
      "balance_loss_mlp": 1.01922464,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 1.7398962768705206,
      "language_loss": 0.72841907,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75046885,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 2.7097504138946533
    },
    {
      "auxiliary_loss_clip": 0.01183761,
      "auxiliary_loss_mlp": 0.0102787,
      "balance_loss_clip": 0.97812295,
      "balance_loss_mlp": 1.02021706,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 2.338815157078354,
      "language_loss": 0.82169974,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84381604,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.6906912326812744
    },
    {
      "auxiliary_loss_clip": 0.01190583,
      "auxiliary_loss_mlp": 0.01035291,
      "balance_loss_clip": 1.01910734,
      "balance_loss_mlp": 1.02775145,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.6417650271897364,
      "language_loss": 0.75790256,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.78016138,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.7309958934783936
    },
    {
      "auxiliary_loss_clip": 0.01171849,
      "auxiliary_loss_mlp": 0.01029883,
      "balance_loss_clip": 0.85861367,
      "balance_loss_mlp": 1.02094233,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 1.7360876832385352,
      "language_loss": 0.76503444,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.7870518,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.8749074935913086
    },
    {
      "auxiliary_loss_clip": 0.01190768,
      "auxiliary_loss_mlp": 0.01034322,
      "balance_loss_clip": 0.9399066,
      "balance_loss_mlp": 1.02599478,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 1.6415459643180637,
      "language_loss": 0.69254279,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.71479362,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 2.756105422973633
    },
    {
      "auxiliary_loss_clip": 0.01185259,
      "auxiliary_loss_mlp": 0.01032289,
      "balance_loss_clip": 1.0548737,
      "balance_loss_mlp": 1.02453399,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 1.4186719695794114,
      "language_loss": 0.75886011,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.78103554,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 2.69443416595459
    },
    {
      "auxiliary_loss_clip": 0.01182194,
      "auxiliary_loss_mlp": 0.01027154,
      "balance_loss_clip": 0.97468269,
      "balance_loss_mlp": 1.01875556,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.5591993014777832,
      "language_loss": 0.72771931,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.74981272,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.7224104404449463
    },
    {
      "auxiliary_loss_clip": 0.01184877,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 1.01436079,
      "balance_loss_mlp": 1.02031469,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 1.9166332704529216,
      "language_loss": 0.74601597,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.76814711,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 2.675802707672119
    },
    {
      "auxiliary_loss_clip": 0.01188095,
      "auxiliary_loss_mlp": 0.01025361,
      "balance_loss_clip": 1.01719522,
      "balance_loss_mlp": 1.01793385,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 1.7217458496515814,
      "language_loss": 0.72444749,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.74658203,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.684577465057373
    },
    {
      "auxiliary_loss_clip": 0.01102122,
      "auxiliary_loss_mlp": 0.01015687,
      "balance_loss_clip": 0.91713393,
      "balance_loss_mlp": 1.01360035,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.7205534073908856,
      "language_loss": 0.61703169,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.63820982,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 3.391031503677368
    },
    {
      "auxiliary_loss_clip": 0.01187282,
      "auxiliary_loss_mlp": 0.01025099,
      "balance_loss_clip": 1.05584335,
      "balance_loss_mlp": 1.01728535,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 1.9821187246806822,
      "language_loss": 0.86634719,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.88847089,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.6970510482788086
    },
    {
      "auxiliary_loss_clip": 0.01181167,
      "auxiliary_loss_mlp": 0.01028319,
      "balance_loss_clip": 0.94018936,
      "balance_loss_mlp": 1.02026629,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 2.0405421854796093,
      "language_loss": 0.86951745,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89161229,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 2.7776873111724854
    },
    {
      "auxiliary_loss_clip": 0.01179175,
      "auxiliary_loss_mlp": 0.01028258,
      "balance_loss_clip": 0.97773898,
      "balance_loss_mlp": 1.01960301,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 1.8918254396483596,
      "language_loss": 0.8153187,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.83739305,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 2.7494237422943115
    },
    {
      "auxiliary_loss_clip": 0.01179661,
      "auxiliary_loss_mlp": 0.0103508,
      "balance_loss_clip": 0.94099188,
      "balance_loss_mlp": 1.02761185,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.8842481387757757,
      "language_loss": 0.79920936,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.82135677,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 3.7780332565307617
    },
    {
      "auxiliary_loss_clip": 0.01184761,
      "auxiliary_loss_mlp": 0.01030468,
      "balance_loss_clip": 0.97617149,
      "balance_loss_mlp": 1.02236211,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 2.0263907985920158,
      "language_loss": 0.759251,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.7814033,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.8263943195343018
    },
    {
      "auxiliary_loss_clip": 0.01178565,
      "auxiliary_loss_mlp": 0.01029116,
      "balance_loss_clip": 0.94134623,
      "balance_loss_mlp": 1.02078331,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.6255049498555465,
      "language_loss": 0.75611192,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.77818871,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.8024768829345703
    },
    {
      "auxiliary_loss_clip": 0.01105061,
      "auxiliary_loss_mlp": 0.00999433,
      "balance_loss_clip": 0.92651051,
      "balance_loss_mlp": 0.99722749,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7298756686079526,
      "language_loss": 0.57713044,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59817541,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.3488729000091553
    },
    {
      "auxiliary_loss_clip": 0.01171682,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 0.97468662,
      "balance_loss_mlp": 1.02067542,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.1570297162149688,
      "language_loss": 0.88353491,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90553385,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.755915880203247
    },
    {
      "auxiliary_loss_clip": 0.01181841,
      "auxiliary_loss_mlp": 0.01023431,
      "balance_loss_clip": 1.01324272,
      "balance_loss_mlp": 1.01573002,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 1.8409035065545512,
      "language_loss": 0.79440677,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.81645948,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 3.6332826614379883
    },
    {
      "auxiliary_loss_clip": 0.01177015,
      "auxiliary_loss_mlp": 0.01032778,
      "balance_loss_clip": 0.93609905,
      "balance_loss_mlp": 1.02514815,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 1.801031336351951,
      "language_loss": 0.81486338,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83696133,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.7352051734924316
    },
    {
      "auxiliary_loss_clip": 0.01187562,
      "auxiliary_loss_mlp": 0.01022843,
      "balance_loss_clip": 0.94095773,
      "balance_loss_mlp": 1.01516604,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 2.0906236990056715,
      "language_loss": 0.75143284,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77353692,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 2.7994067668914795
    },
    {
      "auxiliary_loss_clip": 0.01176486,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 0.9732976,
      "balance_loss_mlp": 1.01576114,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.885172486557838,
      "language_loss": 0.81988895,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.84188521,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 3.5606443881988525
    },
    {
      "auxiliary_loss_clip": 0.01180444,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.01142359,
      "balance_loss_mlp": 1.01835394,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 1.7314074585587502,
      "language_loss": 0.62340665,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.64547628,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.7522900104522705
    },
    {
      "auxiliary_loss_clip": 0.0118682,
      "auxiliary_loss_mlp": 0.01028583,
      "balance_loss_clip": 1.01936519,
      "balance_loss_mlp": 1.02090049,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.4138911512301773,
      "language_loss": 0.83092946,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.85308349,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.6886637210845947
    },
    {
      "auxiliary_loss_clip": 0.01186457,
      "auxiliary_loss_mlp": 0.01028363,
      "balance_loss_clip": 1.01916564,
      "balance_loss_mlp": 1.02016735,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.209906670498662,
      "language_loss": 0.72069752,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.74284571,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.6710915565490723
    },
    {
      "auxiliary_loss_clip": 0.01182249,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 0.98029184,
      "balance_loss_mlp": 1.02021575,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 1.6858630549809825,
      "language_loss": 0.79081267,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81291467,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.708738327026367
    },
    {
      "auxiliary_loss_clip": 0.01181847,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.0523684,
      "balance_loss_mlp": 1.01970935,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.0881353794947954,
      "language_loss": 0.75247639,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.77456498,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.545193672180176
    },
    {
      "auxiliary_loss_clip": 0.01181675,
      "auxiliary_loss_mlp": 0.01022587,
      "balance_loss_clip": 1.01533008,
      "balance_loss_mlp": 1.01530302,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.7914341396823,
      "language_loss": 0.67907536,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70111793,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 2.7367897033691406
    },
    {
      "auxiliary_loss_clip": 0.01177912,
      "auxiliary_loss_mlp": 0.01024803,
      "balance_loss_clip": 0.97620177,
      "balance_loss_mlp": 1.01680422,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 1.7957405479549722,
      "language_loss": 0.68621469,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.70824188,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.6461522579193115
    },
    {
      "auxiliary_loss_clip": 0.01183474,
      "auxiliary_loss_mlp": 0.01031176,
      "balance_loss_clip": 1.01583552,
      "balance_loss_mlp": 1.02319467,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 1.6974181028665312,
      "language_loss": 0.8144871,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83663356,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 2.6927297115325928
    },
    {
      "auxiliary_loss_clip": 0.0118343,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.01751328,
      "balance_loss_mlp": 1.02027607,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 2.013028885925018,
      "language_loss": 0.7583732,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.780487,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.670905590057373
    },
    {
      "auxiliary_loss_clip": 0.01179221,
      "auxiliary_loss_mlp": 0.01025446,
      "balance_loss_clip": 1.05068505,
      "balance_loss_mlp": 1.01818275,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 1.9331872620366677,
      "language_loss": 0.80371135,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82575798,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 2.6210052967071533
    },
    {
      "auxiliary_loss_clip": 0.01180311,
      "auxiliary_loss_mlp": 0.0102579,
      "balance_loss_clip": 1.01416063,
      "balance_loss_mlp": 1.01841736,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 1.7408258944648816,
      "language_loss": 0.67516482,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69722581,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 2.7886176109313965
    },
    {
      "auxiliary_loss_clip": 0.01179079,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.01600671,
      "balance_loss_mlp": 1.02109468,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 1.7465405191546668,
      "language_loss": 0.80647218,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.82855248,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.731779098510742
    },
    {
      "auxiliary_loss_clip": 0.01175917,
      "auxiliary_loss_mlp": 0.01023403,
      "balance_loss_clip": 0.97553778,
      "balance_loss_mlp": 1.01552892,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 2.474481652966291,
      "language_loss": 0.73457122,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.75656444,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.7238004207611084
    },
    {
      "auxiliary_loss_clip": 0.01187349,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.05307317,
      "balance_loss_mlp": 1.01984835,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 2.5494333912374643,
      "language_loss": 0.6778096,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69996643,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 2.637662887573242
    },
    {
      "auxiliary_loss_clip": 0.01182601,
      "auxiliary_loss_mlp": 0.01123586,
      "balance_loss_clip": 0.89914572,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 1.9338353109877333,
      "language_loss": 0.85130632,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.87436819,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.8822524547576904
    },
    {
      "auxiliary_loss_clip": 0.01185171,
      "auxiliary_loss_mlp": 0.0102976,
      "balance_loss_clip": 1.05524135,
      "balance_loss_mlp": 1.02210069,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 1.6700602969881895,
      "language_loss": 0.90413254,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.92628181,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 2.617565155029297
    },
    {
      "auxiliary_loss_clip": 0.01176147,
      "auxiliary_loss_mlp": 0.01025696,
      "balance_loss_clip": 0.97625577,
      "balance_loss_mlp": 1.01792121,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 1.4972282966197832,
      "language_loss": 0.78695357,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.808972,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.7416272163391113
    },
    {
      "auxiliary_loss_clip": 0.01171935,
      "auxiliary_loss_mlp": 0.01123648,
      "balance_loss_clip": 1.0130198,
      "balance_loss_mlp": 0.0,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.1500747761088173,
      "language_loss": 0.91356277,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.93651855,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 3.5729317665100098
    },
    {
      "auxiliary_loss_clip": 0.01190451,
      "auxiliary_loss_mlp": 0.0103041,
      "balance_loss_clip": 0.94123942,
      "balance_loss_mlp": 1.02261972,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 1.9037970757395017,
      "language_loss": 0.66399884,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68620753,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 3.6913881301879883
    },
    {
      "auxiliary_loss_clip": 0.01172571,
      "auxiliary_loss_mlp": 0.01023959,
      "balance_loss_clip": 1.01307893,
      "balance_loss_mlp": 1.01668751,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 1.626014871802519,
      "language_loss": 0.65583676,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.67780209,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.679292678833008
    },
    {
      "auxiliary_loss_clip": 0.01172387,
      "auxiliary_loss_mlp": 0.01024698,
      "balance_loss_clip": 0.97576988,
      "balance_loss_mlp": 1.01682973,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 2.2181760865826208,
      "language_loss": 0.62420952,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.64618039,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.6789796352386475
    },
    {
      "auxiliary_loss_clip": 0.01186894,
      "auxiliary_loss_mlp": 0.0103137,
      "balance_loss_clip": 1.01786053,
      "balance_loss_mlp": 1.02360034,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 1.853805409565906,
      "language_loss": 0.67917848,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.70136106,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.6861181259155273
    },
    {
      "auxiliary_loss_clip": 0.01181661,
      "auxiliary_loss_mlp": 0.01031749,
      "balance_loss_clip": 0.93873096,
      "balance_loss_mlp": 1.02368426,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 1.678432039846757,
      "language_loss": 0.79173154,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81386566,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 3.6959245204925537
    },
    {
      "auxiliary_loss_clip": 0.01182211,
      "auxiliary_loss_mlp": 0.01028228,
      "balance_loss_clip": 1.01286197,
      "balance_loss_mlp": 1.02041411,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 1.7381677229104024,
      "language_loss": 0.68479085,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70689523,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.6533095836639404
    },
    {
      "auxiliary_loss_clip": 0.01182959,
      "auxiliary_loss_mlp": 0.01030878,
      "balance_loss_clip": 0.97910643,
      "balance_loss_mlp": 1.02297997,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 1.6973311803780784,
      "language_loss": 0.77276015,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79489851,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.729245662689209
    },
    {
      "auxiliary_loss_clip": 0.01173751,
      "auxiliary_loss_mlp": 0.01029057,
      "balance_loss_clip": 0.97527158,
      "balance_loss_mlp": 1.02093852,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 1.9695889287800579,
      "language_loss": 0.74008518,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.76211327,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 3.6574535369873047
    },
    {
      "auxiliary_loss_clip": 0.01179342,
      "auxiliary_loss_mlp": 0.01025386,
      "balance_loss_clip": 0.93723893,
      "balance_loss_mlp": 1.01698136,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 1.8552474305021778,
      "language_loss": 0.79496706,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.81701434,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.8359010219573975
    },
    {
      "auxiliary_loss_clip": 0.01181058,
      "auxiliary_loss_mlp": 0.01032956,
      "balance_loss_clip": 0.93637514,
      "balance_loss_mlp": 1.0252285,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 3.0987232662289625,
      "language_loss": 0.7792592,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80139941,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.8177385330200195
    },
    {
      "auxiliary_loss_clip": 0.01179906,
      "auxiliary_loss_mlp": 0.01034229,
      "balance_loss_clip": 1.01477563,
      "balance_loss_mlp": 1.02661395,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.6726819906799482,
      "language_loss": 0.81270522,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.83484662,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.717115879058838
    },
    {
      "auxiliary_loss_clip": 0.01098211,
      "auxiliary_loss_mlp": 0.01015145,
      "balance_loss_clip": 0.87407261,
      "balance_loss_mlp": 1.01315391,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7558983276301114,
      "language_loss": 0.60465503,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62578857,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.4622533321380615
    },
    {
      "auxiliary_loss_clip": 0.01186395,
      "auxiliary_loss_mlp": 0.01124144,
      "balance_loss_clip": 0.97832501,
      "balance_loss_mlp": 0.0,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.639077474985095,
      "language_loss": 0.77496946,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.79807484,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.932847738265991
    },
    {
      "auxiliary_loss_clip": 0.01167588,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 0.81779414,
      "balance_loss_mlp": 1.01905584,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.9055796148469946,
      "language_loss": 0.7692821,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79122281,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 3.0103278160095215
    },
    {
      "auxiliary_loss_clip": 0.01181252,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 1.01688373,
      "balance_loss_mlp": 1.022686,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 2.233120930261395,
      "language_loss": 0.77069372,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.79280484,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.942396879196167
    },
    {
      "auxiliary_loss_clip": 0.01189974,
      "auxiliary_loss_mlp": 0.01123918,
      "balance_loss_clip": 1.01734805,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 1.8451614060838075,
      "language_loss": 0.69796824,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.72110713,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.641726493835449
    },
    {
      "auxiliary_loss_clip": 0.011814,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.05306792,
      "balance_loss_mlp": 1.02241611,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 1.6405192321112863,
      "language_loss": 0.8347221,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.8568294,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 2.687281370162964
    },
    {
      "auxiliary_loss_clip": 0.01186109,
      "auxiliary_loss_mlp": 0.01030878,
      "balance_loss_clip": 0.97983432,
      "balance_loss_mlp": 1.02326608,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.58946689460216,
      "language_loss": 0.79656994,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.81873977,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 2.8160016536712646
    },
    {
      "auxiliary_loss_clip": 0.0118687,
      "auxiliary_loss_mlp": 0.01028416,
      "balance_loss_clip": 0.93772054,
      "balance_loss_mlp": 1.02037501,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 2.0568851971076194,
      "language_loss": 0.79647416,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.81862706,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 2.795614719390869
    },
    {
      "auxiliary_loss_clip": 0.01177876,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 0.97388828,
      "balance_loss_mlp": 1.0196712,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 1.7521937898700994,
      "language_loss": 0.76027626,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78233165,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 2.705929756164551
    },
    {
      "auxiliary_loss_clip": 0.01183612,
      "auxiliary_loss_mlp": 0.01029824,
      "balance_loss_clip": 0.97806638,
      "balance_loss_mlp": 1.0221591,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 1.8915702184828371,
      "language_loss": 0.8179794,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.84011376,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.7552804946899414
    },
    {
      "auxiliary_loss_clip": 0.0118775,
      "auxiliary_loss_mlp": 0.01032811,
      "balance_loss_clip": 0.89978743,
      "balance_loss_mlp": 1.02459192,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 2.0872169749077556,
      "language_loss": 0.78883994,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.81104553,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 2.8426802158355713
    },
    {
      "auxiliary_loss_clip": 0.01184982,
      "auxiliary_loss_mlp": 0.01029932,
      "balance_loss_clip": 1.01637411,
      "balance_loss_mlp": 1.0221653,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.623188437974138,
      "language_loss": 0.80460703,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.82675624,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.702329397201538
    },
    {
      "auxiliary_loss_clip": 0.01181528,
      "auxiliary_loss_mlp": 0.01035754,
      "balance_loss_clip": 0.93877786,
      "balance_loss_mlp": 1.0272007,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 2.002040147750967,
      "language_loss": 0.74993491,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.77210772,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 2.7306578159332275
    },
    {
      "auxiliary_loss_clip": 0.01182256,
      "auxiliary_loss_mlp": 0.01025567,
      "balance_loss_clip": 0.97567594,
      "balance_loss_mlp": 1.01824427,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 1.7379835844331653,
      "language_loss": 0.79896235,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82104057,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 2.739915609359741
    },
    {
      "auxiliary_loss_clip": 0.01181215,
      "auxiliary_loss_mlp": 0.01023365,
      "balance_loss_clip": 0.90079165,
      "balance_loss_mlp": 1.01581597,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 1.8438382723805489,
      "language_loss": 0.8168211,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83886695,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 4.752245187759399
    },
    {
      "auxiliary_loss_clip": 0.01176857,
      "auxiliary_loss_mlp": 0.01027117,
      "balance_loss_clip": 0.97337878,
      "balance_loss_mlp": 1.01964903,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.625175873063264,
      "language_loss": 0.7661438,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.78818345,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.825624465942383
    },
    {
      "auxiliary_loss_clip": 0.01181046,
      "auxiliary_loss_mlp": 0.01023229,
      "balance_loss_clip": 0.93535209,
      "balance_loss_mlp": 1.01516461,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.233836863702708,
      "language_loss": 0.68092692,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.70296973,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.7911128997802734
    },
    {
      "auxiliary_loss_clip": 0.01185877,
      "auxiliary_loss_mlp": 0.01025107,
      "balance_loss_clip": 1.05702925,
      "balance_loss_mlp": 1.01824915,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 1.9376829785432201,
      "language_loss": 0.84504938,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.86715925,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.696614980697632
    },
    {
      "auxiliary_loss_clip": 0.01189858,
      "auxiliary_loss_mlp": 0.01029144,
      "balance_loss_clip": 1.01627588,
      "balance_loss_mlp": 1.02067983,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 2.5479944808969237,
      "language_loss": 0.66570121,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.68789124,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.654890298843384
    },
    {
      "auxiliary_loss_clip": 0.01089887,
      "auxiliary_loss_mlp": 0.0100325,
      "balance_loss_clip": 1.02408433,
      "balance_loss_mlp": 1.00127113,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.8293281087112214,
      "language_loss": 0.55560112,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57653248,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 4.112024307250977
    },
    {
      "auxiliary_loss_clip": 0.01089664,
      "auxiliary_loss_mlp": 0.01003853,
      "balance_loss_clip": 0.98709649,
      "balance_loss_mlp": 1.00185013,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.7660591513436,
      "language_loss": 0.61232287,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63325804,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.1536922454833984
    },
    {
      "auxiliary_loss_clip": 0.01171264,
      "auxiliary_loss_mlp": 0.01029479,
      "balance_loss_clip": 0.90045649,
      "balance_loss_mlp": 1.02146173,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 2.302432576615666,
      "language_loss": 0.81917417,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.84118158,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.7516438961029053
    },
    {
      "auxiliary_loss_clip": 0.01103094,
      "auxiliary_loss_mlp": 0.00999345,
      "balance_loss_clip": 0.92473114,
      "balance_loss_mlp": 0.99716395,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.8363095351618923,
      "language_loss": 0.58873427,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.60975862,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 4.110677719116211
    },
    {
      "auxiliary_loss_clip": 0.01186937,
      "auxiliary_loss_mlp": 0.01034863,
      "balance_loss_clip": 1.02014267,
      "balance_loss_mlp": 1.0273056,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 1.6066414906206221,
      "language_loss": 0.8100487,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83226669,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 2.6920993328094482
    },
    {
      "auxiliary_loss_clip": 0.0118567,
      "auxiliary_loss_mlp": 0.01027852,
      "balance_loss_clip": 0.98184228,
      "balance_loss_mlp": 1.01978719,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 1.6581786891644894,
      "language_loss": 0.77361012,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.79574537,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.7088704109191895
    },
    {
      "auxiliary_loss_clip": 0.01180162,
      "auxiliary_loss_mlp": 0.01029142,
      "balance_loss_clip": 0.82407844,
      "balance_loss_mlp": 1.02047586,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.5523560213627727,
      "language_loss": 0.81675291,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.83884597,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 2.9472196102142334
    },
    {
      "auxiliary_loss_clip": 0.01186738,
      "auxiliary_loss_mlp": 0.01030873,
      "balance_loss_clip": 0.94120413,
      "balance_loss_mlp": 1.02299297,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 1.710166150888007,
      "language_loss": 0.81898177,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84115779,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 3.101139783859253
    },
    {
      "auxiliary_loss_clip": 0.0118826,
      "auxiliary_loss_mlp": 0.01026954,
      "balance_loss_clip": 1.01773524,
      "balance_loss_mlp": 1.01809645,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 1.9132178074290467,
      "language_loss": 0.87828714,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.90043926,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.687917947769165
    },
    {
      "auxiliary_loss_clip": 0.01174848,
      "auxiliary_loss_mlp": 0.01124607,
      "balance_loss_clip": 0.93483424,
      "balance_loss_mlp": 0.0,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.5243591269866772,
      "language_loss": 0.85288048,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.87587506,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 2.7880496978759766
    },
    {
      "auxiliary_loss_clip": 0.01189741,
      "auxiliary_loss_mlp": 0.01124144,
      "balance_loss_clip": 0.97643828,
      "balance_loss_mlp": 0.0,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 4.40086187876193,
      "language_loss": 0.82693613,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.85007501,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.7392804622650146
    },
    {
      "auxiliary_loss_clip": 0.01181868,
      "auxiliary_loss_mlp": 0.01027297,
      "balance_loss_clip": 0.97835726,
      "balance_loss_mlp": 1.01948881,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 3.8772601303098977,
      "language_loss": 0.65938187,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.68147349,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.63464617729187
    },
    {
      "auxiliary_loss_clip": 0.01188332,
      "auxiliary_loss_mlp": 0.01028795,
      "balance_loss_clip": 1.01797295,
      "balance_loss_mlp": 1.02074218,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 1.8443292970383252,
      "language_loss": 0.64837712,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67054838,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 2.741594076156616
    },
    {
      "auxiliary_loss_clip": 0.01185226,
      "auxiliary_loss_mlp": 0.01028094,
      "balance_loss_clip": 1.0146457,
      "balance_loss_mlp": 1.0204711,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 1.6631241047681982,
      "language_loss": 0.76677001,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78890318,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 2.7270405292510986
    },
    {
      "auxiliary_loss_clip": 0.0118689,
      "auxiliary_loss_mlp": 0.01038945,
      "balance_loss_clip": 1.05418897,
      "balance_loss_mlp": 1.03017688,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 1.8519298920933012,
      "language_loss": 0.77306545,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79532373,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 2.6591060161590576
    },
    {
      "auxiliary_loss_clip": 0.01178613,
      "auxiliary_loss_mlp": 0.01026106,
      "balance_loss_clip": 0.90128392,
      "balance_loss_mlp": 1.01789784,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 1.7902384330988332,
      "language_loss": 0.87308812,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89513528,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 2.818284749984741
    },
    {
      "auxiliary_loss_clip": 0.01186189,
      "auxiliary_loss_mlp": 0.01025874,
      "balance_loss_clip": 0.90073776,
      "balance_loss_mlp": 1.0181253,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 2.1326133331514163,
      "language_loss": 0.93034434,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95246506,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.7369704246520996
    },
    {
      "auxiliary_loss_clip": 0.0118833,
      "auxiliary_loss_mlp": 0.01123725,
      "balance_loss_clip": 1.01788807,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 2.105046232245987,
      "language_loss": 0.65467417,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.67779469,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.7293717861175537
    },
    {
      "auxiliary_loss_clip": 0.01180138,
      "auxiliary_loss_mlp": 0.01124369,
      "balance_loss_clip": 0.94063354,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.3554236020413275,
      "language_loss": 0.94412887,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.96717393,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 2.738337516784668
    },
    {
      "auxiliary_loss_clip": 0.01183365,
      "auxiliary_loss_mlp": 0.01028683,
      "balance_loss_clip": 1.01535296,
      "balance_loss_mlp": 1.02098465,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 2.2575908857764273,
      "language_loss": 0.88255417,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90467465,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 2.7711431980133057
    },
    {
      "auxiliary_loss_clip": 0.0118321,
      "auxiliary_loss_mlp": 0.0102528,
      "balance_loss_clip": 1.01758087,
      "balance_loss_mlp": 1.01761198,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 3.06544659674576,
      "language_loss": 0.77342898,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.79551387,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 2.704608201980591
    },
    {
      "auxiliary_loss_clip": 0.01181784,
      "auxiliary_loss_mlp": 0.01031474,
      "balance_loss_clip": 0.97506005,
      "balance_loss_mlp": 1.0236001,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 2.3647631327451313,
      "language_loss": 0.85033154,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87246412,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 4.108766794204712
    },
    {
      "auxiliary_loss_clip": 0.01184435,
      "auxiliary_loss_mlp": 0.01025412,
      "balance_loss_clip": 1.01830196,
      "balance_loss_mlp": 1.01791334,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 1.7429632600966787,
      "language_loss": 0.80015266,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.82225114,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 3.669327735900879
    },
    {
      "auxiliary_loss_clip": 0.0119348,
      "auxiliary_loss_mlp": 0.01026327,
      "balance_loss_clip": 0.93947053,
      "balance_loss_mlp": 1.0179168,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 1.9512174181672175,
      "language_loss": 0.90149701,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.92369509,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.824557065963745
    },
    {
      "auxiliary_loss_clip": 0.01181184,
      "auxiliary_loss_mlp": 0.01026078,
      "balance_loss_clip": 0.89823067,
      "balance_loss_mlp": 1.01835299,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 3.4642596981334375,
      "language_loss": 0.75328833,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.77536094,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.8606865406036377
    },
    {
      "auxiliary_loss_clip": 0.01094187,
      "auxiliary_loss_mlp": 0.01004222,
      "balance_loss_clip": 0.91073513,
      "balance_loss_mlp": 1.0022788,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 1.1187131942463926,
      "language_loss": 0.60919964,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.6301837,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.149294137954712
    },
    {
      "auxiliary_loss_clip": 0.01186209,
      "auxiliary_loss_mlp": 0.01024835,
      "balance_loss_clip": 1.01648545,
      "balance_loss_mlp": 1.01665163,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 2.0741073503794074,
      "language_loss": 0.81171656,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83382702,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.6656668186187744
    },
    {
      "auxiliary_loss_clip": 0.01177577,
      "auxiliary_loss_mlp": 0.01029119,
      "balance_loss_clip": 1.01368785,
      "balance_loss_mlp": 1.02066112,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 1.7644712520038264,
      "language_loss": 0.68118042,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70324737,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 3.627284049987793
    },
    {
      "auxiliary_loss_clip": 0.0117347,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 0.7862305,
      "balance_loss_mlp": 1.02211881,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 2.0494299085361916,
      "language_loss": 0.80073869,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.82277822,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 2.9785990715026855
    },
    {
      "auxiliary_loss_clip": 0.01185303,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 0.97804934,
      "balance_loss_mlp": 1.02018487,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.6522923519936938,
      "language_loss": 0.80082905,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82296312,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 4.191251277923584
    },
    {
      "auxiliary_loss_clip": 0.01186452,
      "auxiliary_loss_mlp": 0.01028314,
      "balance_loss_clip": 1.01315284,
      "balance_loss_mlp": 1.02024961,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.4663146345769913,
      "language_loss": 0.87210977,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89425743,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.7000906467437744
    },
    {
      "auxiliary_loss_clip": 0.01179865,
      "auxiliary_loss_mlp": 0.0103012,
      "balance_loss_clip": 1.01388788,
      "balance_loss_mlp": 1.02275252,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 1.7916700857020165,
      "language_loss": 0.81307459,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.8351745,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.696242332458496
    },
    {
      "auxiliary_loss_clip": 0.01178381,
      "auxiliary_loss_mlp": 0.01025783,
      "balance_loss_clip": 0.93531954,
      "balance_loss_mlp": 1.01835048,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 1.8826273944498801,
      "language_loss": 0.81102872,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.83307028,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 2.697474718093872
    },
    {
      "auxiliary_loss_clip": 0.01181909,
      "auxiliary_loss_mlp": 0.01032183,
      "balance_loss_clip": 0.94211638,
      "balance_loss_mlp": 1.02453542,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.575704158628241,
      "language_loss": 0.6620512,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68419218,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.695726156234741
    },
    {
      "auxiliary_loss_clip": 0.0119156,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.01556897,
      "balance_loss_mlp": 1.02013469,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 2.0051801769662605,
      "language_loss": 0.74818623,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.77038687,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.612287998199463
    },
    {
      "auxiliary_loss_clip": 0.01171865,
      "auxiliary_loss_mlp": 0.01029717,
      "balance_loss_clip": 0.93719065,
      "balance_loss_mlp": 1.02162242,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.8196519979761998,
      "language_loss": 0.87794018,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.89995599,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.673452615737915
    },
    {
      "auxiliary_loss_clip": 0.01089191,
      "auxiliary_loss_mlp": 0.01002785,
      "balance_loss_clip": 0.98527336,
      "balance_loss_mlp": 1.00081778,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.8166284628052796,
      "language_loss": 0.53416699,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.55508673,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.289127826690674
    },
    {
      "auxiliary_loss_clip": 0.01181735,
      "auxiliary_loss_mlp": 0.01031119,
      "balance_loss_clip": 0.97832859,
      "balance_loss_mlp": 1.02243471,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.606292905015003,
      "language_loss": 0.69627738,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.71840596,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.623948097229004
    },
    {
      "auxiliary_loss_clip": 0.01178416,
      "auxiliary_loss_mlp": 0.01124193,
      "balance_loss_clip": 0.97829288,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 1.7125916560748982,
      "language_loss": 0.65962172,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.68264782,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 2.671678304672241
    },
    {
      "auxiliary_loss_clip": 0.01186241,
      "auxiliary_loss_mlp": 0.01023481,
      "balance_loss_clip": 0.8262971,
      "balance_loss_mlp": 1.01506519,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 2.404922049567735,
      "language_loss": 0.63657314,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.65867043,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 2.909273862838745
    },
    {
      "auxiliary_loss_clip": 0.0118355,
      "auxiliary_loss_mlp": 0.010265,
      "balance_loss_clip": 0.93957335,
      "balance_loss_mlp": 1.01794672,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 3.1295667410986847,
      "language_loss": 0.75546128,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.77756178,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 3.2706127166748047
    },
    {
      "auxiliary_loss_clip": 0.01181355,
      "auxiliary_loss_mlp": 0.01023846,
      "balance_loss_clip": 0.89760202,
      "balance_loss_mlp": 1.01608002,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 2.9042307159306824,
      "language_loss": 0.70045257,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.72250462,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 2.8505144119262695
    },
    {
      "auxiliary_loss_clip": 0.01184262,
      "auxiliary_loss_mlp": 0.01027486,
      "balance_loss_clip": 1.01729989,
      "balance_loss_mlp": 1.01962399,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 2.0298283432202644,
      "language_loss": 0.83795977,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.86007726,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 2.668567657470703
    },
    {
      "auxiliary_loss_clip": 0.01184924,
      "auxiliary_loss_mlp": 0.01033055,
      "balance_loss_clip": 1.05487633,
      "balance_loss_mlp": 1.02535105,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 1.6617780828272366,
      "language_loss": 0.75283313,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77501297,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.6219189167022705
    },
    {
      "auxiliary_loss_clip": 0.01180254,
      "auxiliary_loss_mlp": 0.01033081,
      "balance_loss_clip": 1.01485014,
      "balance_loss_mlp": 1.0244565,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 1.71189641964567,
      "language_loss": 0.77188396,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79401726,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 2.685880422592163
    },
    {
      "auxiliary_loss_clip": 0.01183996,
      "auxiliary_loss_mlp": 0.01027317,
      "balance_loss_clip": 0.9389298,
      "balance_loss_mlp": 1.01857877,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 1.8671233898094108,
      "language_loss": 0.81647098,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.83858413,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.7238776683807373
    },
    {
      "auxiliary_loss_clip": 0.01181683,
      "auxiliary_loss_mlp": 0.01028667,
      "balance_loss_clip": 0.9756465,
      "balance_loss_mlp": 1.02117455,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 2.4289679500521926,
      "language_loss": 0.8841393,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90624279,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 2.7915492057800293
    },
    {
      "auxiliary_loss_clip": 0.01178865,
      "auxiliary_loss_mlp": 0.01026215,
      "balance_loss_clip": 1.01396728,
      "balance_loss_mlp": 1.01852655,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.6214306322162644,
      "language_loss": 0.76729697,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.78934777,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 4.853153944015503
    },
    {
      "auxiliary_loss_clip": 0.01184662,
      "auxiliary_loss_mlp": 0.01024697,
      "balance_loss_clip": 1.01559472,
      "balance_loss_mlp": 1.01681423,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.4891972521630263,
      "language_loss": 0.65513104,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67722464,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.720045566558838
    },
    {
      "auxiliary_loss_clip": 0.01189075,
      "auxiliary_loss_mlp": 0.01030517,
      "balance_loss_clip": 1.01534462,
      "balance_loss_mlp": 1.02247643,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.1704771106231067,
      "language_loss": 0.87107944,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89327538,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.6306116580963135
    },
    {
      "auxiliary_loss_clip": 0.01186324,
      "auxiliary_loss_mlp": 0.01029703,
      "balance_loss_clip": 1.05410361,
      "balance_loss_mlp": 1.02154899,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 1.910914770297639,
      "language_loss": 0.79187739,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81403768,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.6045942306518555
    },
    {
      "auxiliary_loss_clip": 0.01171675,
      "auxiliary_loss_mlp": 0.01034771,
      "balance_loss_clip": 0.90095812,
      "balance_loss_mlp": 1.02665925,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 1.9442464803437143,
      "language_loss": 0.67842376,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.70048821,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.8039252758026123
    },
    {
      "auxiliary_loss_clip": 0.01172138,
      "auxiliary_loss_mlp": 0.01030311,
      "balance_loss_clip": 0.97378743,
      "balance_loss_mlp": 1.0222944,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 2.061429184227631,
      "language_loss": 0.64376402,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66578853,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.608623504638672
    },
    {
      "auxiliary_loss_clip": 0.01185162,
      "auxiliary_loss_mlp": 0.01030976,
      "balance_loss_clip": 1.05400479,
      "balance_loss_mlp": 1.02272058,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 2.2773894973666127,
      "language_loss": 0.83811617,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.86027753,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 3.5107758045196533
    },
    {
      "auxiliary_loss_clip": 0.01189531,
      "auxiliary_loss_mlp": 0.0103065,
      "balance_loss_clip": 0.97881168,
      "balance_loss_mlp": 1.02247262,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.043705757102316,
      "language_loss": 0.75280637,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.7750082,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.795139789581299
    },
    {
      "auxiliary_loss_clip": 0.01185008,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.0542767,
      "balance_loss_mlp": 1.01928377,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.758639624824342,
      "language_loss": 0.73319292,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75531453,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 3.645063638687134
    },
    {
      "auxiliary_loss_clip": 0.01187672,
      "auxiliary_loss_mlp": 0.01030279,
      "balance_loss_clip": 1.05575466,
      "balance_loss_mlp": 1.02169013,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 2.499656825063409,
      "language_loss": 0.78889835,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.81107783,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.6065306663513184
    },
    {
      "auxiliary_loss_clip": 0.0118648,
      "auxiliary_loss_mlp": 0.01124192,
      "balance_loss_clip": 1.01703942,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.3376905694241974,
      "language_loss": 0.77219176,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.79529846,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.6345746517181396
    },
    {
      "auxiliary_loss_clip": 0.0118525,
      "auxiliary_loss_mlp": 0.01026038,
      "balance_loss_clip": 0.975999,
      "balance_loss_mlp": 1.01759183,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 1.9891909229242413,
      "language_loss": 0.89716196,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.91927481,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.6725754737854004
    },
    {
      "auxiliary_loss_clip": 0.01087608,
      "auxiliary_loss_mlp": 0.01003386,
      "balance_loss_clip": 0.94587958,
      "balance_loss_mlp": 1.00135922,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7874515006927079,
      "language_loss": 0.57374156,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.59465146,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.280794143676758
    },
    {
      "auxiliary_loss_clip": 0.01181828,
      "auxiliary_loss_mlp": 0.01024118,
      "balance_loss_clip": 0.93829083,
      "balance_loss_mlp": 1.01647687,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 2.121473599068382,
      "language_loss": 0.78524023,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.80729967,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.728649616241455
    },
    {
      "auxiliary_loss_clip": 0.01174491,
      "auxiliary_loss_mlp": 0.01030891,
      "balance_loss_clip": 0.85998952,
      "balance_loss_mlp": 1.02313638,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 1.7139077657812427,
      "language_loss": 0.86380041,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.88585418,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.884364366531372
    },
    {
      "auxiliary_loss_clip": 0.01184187,
      "auxiliary_loss_mlp": 0.01029883,
      "balance_loss_clip": 1.01335418,
      "balance_loss_mlp": 1.02247453,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 2.153724679409345,
      "language_loss": 0.74703264,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.76917338,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 2.760944128036499
    },
    {
      "auxiliary_loss_clip": 0.0117194,
      "auxiliary_loss_mlp": 0.0102643,
      "balance_loss_clip": 0.89572573,
      "balance_loss_mlp": 1.01836598,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.182209717711598,
      "language_loss": 0.71447432,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.73645806,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.7151060104370117
    },
    {
      "auxiliary_loss_clip": 0.01186665,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 0.90224552,
      "balance_loss_mlp": 1.01932192,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 2.3792055862777586,
      "language_loss": 0.61381292,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.635957,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 2.7860217094421387
    },
    {
      "auxiliary_loss_clip": 0.01180943,
      "auxiliary_loss_mlp": 0.01026179,
      "balance_loss_clip": 0.97785628,
      "balance_loss_mlp": 1.01822746,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 2.0764419508380034,
      "language_loss": 0.7485503,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.77062154,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 2.6534831523895264
    },
    {
      "auxiliary_loss_clip": 0.01188004,
      "auxiliary_loss_mlp": 0.01030727,
      "balance_loss_clip": 1.01549864,
      "balance_loss_mlp": 1.02228701,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 1.7361766271921892,
      "language_loss": 0.80767912,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.82986647,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 2.61525821685791
    },
    {
      "auxiliary_loss_clip": 0.01171493,
      "auxiliary_loss_mlp": 0.01022909,
      "balance_loss_clip": 0.89570236,
      "balance_loss_mlp": 1.01499391,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 2.29217240152575,
      "language_loss": 0.81876236,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84070647,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 2.7404251098632812
    },
    {
      "auxiliary_loss_clip": 0.01180468,
      "auxiliary_loss_mlp": 0.0102636,
      "balance_loss_clip": 0.93664432,
      "balance_loss_mlp": 1.01834297,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 1.922622799526427,
      "language_loss": 0.61299288,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.63506114,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 2.6807310581207275
    },
    {
      "auxiliary_loss_clip": 0.01185955,
      "auxiliary_loss_mlp": 0.01023327,
      "balance_loss_clip": 0.862809,
      "balance_loss_mlp": 1.01537013,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 2.0477310473106187,
      "language_loss": 0.872648,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.89474082,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.8751537799835205
    },
    {
      "auxiliary_loss_clip": 0.01182132,
      "auxiliary_loss_mlp": 0.01124319,
      "balance_loss_clip": 0.93716055,
      "balance_loss_mlp": 0.0,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 2.2959048862900397,
      "language_loss": 0.74958515,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.77264965,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 2.7062220573425293
    },
    {
      "auxiliary_loss_clip": 0.01171894,
      "auxiliary_loss_mlp": 0.01030331,
      "balance_loss_clip": 0.93342876,
      "balance_loss_mlp": 1.02208734,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.3322563720244265,
      "language_loss": 0.83932018,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.86134243,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.6975765228271484
    },
    {
      "auxiliary_loss_clip": 0.01184999,
      "auxiliary_loss_mlp": 0.01030191,
      "balance_loss_clip": 0.97373605,
      "balance_loss_mlp": 1.02225173,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 1.8854594526950803,
      "language_loss": 0.72887683,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75102878,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 3.6851584911346436
    },
    {
      "auxiliary_loss_clip": 0.01185952,
      "auxiliary_loss_mlp": 0.01034514,
      "balance_loss_clip": 1.05424452,
      "balance_loss_mlp": 1.02667022,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.6704429363393547,
      "language_loss": 0.73400015,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75620484,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 2.725320816040039
    },
    {
      "auxiliary_loss_clip": 0.01184643,
      "auxiliary_loss_mlp": 0.01024439,
      "balance_loss_clip": 0.90035075,
      "balance_loss_mlp": 1.01643968,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 2.029772920237369,
      "language_loss": 0.84336865,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.8654595,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 3.7611265182495117
    },
    {
      "auxiliary_loss_clip": 0.01186006,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.01343942,
      "balance_loss_mlp": 1.01971745,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 1.8220413709400145,
      "language_loss": 0.80993122,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.83207065,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.6626741886138916
    },
    {
      "auxiliary_loss_clip": 0.01177191,
      "auxiliary_loss_mlp": 0.0102886,
      "balance_loss_clip": 1.01284432,
      "balance_loss_mlp": 1.02084935,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 1.775714839284276,
      "language_loss": 0.78932452,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81138504,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.725003242492676
    },
    {
      "auxiliary_loss_clip": 0.01177333,
      "auxiliary_loss_mlp": 0.01026657,
      "balance_loss_clip": 1.01313698,
      "balance_loss_mlp": 1.0184797,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.7080719784799645,
      "language_loss": 0.77376318,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.79580307,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.7051844596862793
    },
    {
      "auxiliary_loss_clip": 0.01187774,
      "auxiliary_loss_mlp": 0.01029082,
      "balance_loss_clip": 1.05347204,
      "balance_loss_mlp": 1.02066016,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 1.95800759404508,
      "language_loss": 0.79090226,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81307083,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.6864259243011475
    },
    {
      "auxiliary_loss_clip": 0.01181592,
      "auxiliary_loss_mlp": 0.01028887,
      "balance_loss_clip": 1.01510012,
      "balance_loss_mlp": 1.02092361,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 1.9800928605224248,
      "language_loss": 0.84788501,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.86998975,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 4.022350549697876
    },
    {
      "auxiliary_loss_clip": 0.01181158,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.01592588,
      "balance_loss_mlp": 1.02191222,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.598718348933168,
      "language_loss": 0.80549085,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.82759821,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.668365716934204
    },
    {
      "auxiliary_loss_clip": 0.01185859,
      "auxiliary_loss_mlp": 0.0102516,
      "balance_loss_clip": 1.01518321,
      "balance_loss_mlp": 1.01748872,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 1.3807477129000632,
      "language_loss": 0.7707634,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.79287356,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 3.63617205619812
    },
    {
      "auxiliary_loss_clip": 0.01187896,
      "auxiliary_loss_mlp": 0.01027424,
      "balance_loss_clip": 0.94045061,
      "balance_loss_mlp": 1.01918626,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 2.093601471342815,
      "language_loss": 0.72527158,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.74742472,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.6949081420898438
    },
    {
      "auxiliary_loss_clip": 0.0118245,
      "auxiliary_loss_mlp": 0.01032898,
      "balance_loss_clip": 1.0163486,
      "balance_loss_mlp": 1.02514958,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 1.7401254879264967,
      "language_loss": 0.75295824,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77511168,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.657477617263794
    },
    {
      "auxiliary_loss_clip": 0.01082253,
      "auxiliary_loss_mlp": 0.01002734,
      "balance_loss_clip": 0.98251355,
      "balance_loss_mlp": 1.00068367,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.7599659521166064,
      "language_loss": 0.56558371,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.58643365,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.206023693084717
    },
    {
      "auxiliary_loss_clip": 0.01181876,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 1.01666832,
      "balance_loss_mlp": 1.01929045,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 1.742424110912525,
      "language_loss": 0.69828135,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.72037005,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.683504343032837
    },
    {
      "auxiliary_loss_clip": 0.01183319,
      "auxiliary_loss_mlp": 0.01032045,
      "balance_loss_clip": 0.94020641,
      "balance_loss_mlp": 1.02336097,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 2.0801434522058693,
      "language_loss": 0.73600543,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.75815916,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.7357757091522217
    },
    {
      "auxiliary_loss_clip": 0.01183519,
      "auxiliary_loss_mlp": 0.01029011,
      "balance_loss_clip": 1.01473832,
      "balance_loss_mlp": 1.02048159,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 2.0986580382834013,
      "language_loss": 0.78658175,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.808707,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.665936231613159
    },
    {
      "auxiliary_loss_clip": 0.01183622,
      "auxiliary_loss_mlp": 0.01027817,
      "balance_loss_clip": 0.9767468,
      "balance_loss_mlp": 1.02009797,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.3120767892585667,
      "language_loss": 0.8287909,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.8509053,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.666853666305542
    },
    {
      "auxiliary_loss_clip": 0.01173199,
      "auxiliary_loss_mlp": 0.01027294,
      "balance_loss_clip": 0.97326565,
      "balance_loss_mlp": 1.01909244,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 2.185875402908665,
      "language_loss": 0.81037009,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83237505,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 2.6983795166015625
    },
    {
      "auxiliary_loss_clip": 0.0119034,
      "auxiliary_loss_mlp": 0.01124492,
      "balance_loss_clip": 1.0552218,
      "balance_loss_mlp": 0.0,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 2.7107593309170306,
      "language_loss": 0.76747298,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.79062128,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 2.5598413944244385
    },
    {
      "auxiliary_loss_clip": 0.01176818,
      "auxiliary_loss_mlp": 0.01031346,
      "balance_loss_clip": 0.90380889,
      "balance_loss_mlp": 1.02297735,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 2.2019372221947964,
      "language_loss": 0.74067378,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.76275539,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 2.733046770095825
    },
    {
      "auxiliary_loss_clip": 0.01187837,
      "auxiliary_loss_mlp": 0.01032956,
      "balance_loss_clip": 1.01689637,
      "balance_loss_mlp": 1.02414608,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 2.9161713640824747,
      "language_loss": 0.68505812,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70726609,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 2.6342098712921143
    },
    {
      "auxiliary_loss_clip": 0.01183832,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 0.97574341,
      "balance_loss_mlp": 1.0191747,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.736698109184975,
      "language_loss": 0.63885629,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.6609655,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 2.7033612728118896
    },
    {
      "auxiliary_loss_clip": 0.01177918,
      "auxiliary_loss_mlp": 0.01031717,
      "balance_loss_clip": 0.93936265,
      "balance_loss_mlp": 1.02378941,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 1.630308015307728,
      "language_loss": 0.71977055,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.74186683,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 2.674546718597412
    },
    {
      "auxiliary_loss_clip": 0.01169631,
      "auxiliary_loss_mlp": 0.01029964,
      "balance_loss_clip": 0.8985377,
      "balance_loss_mlp": 1.02211368,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.6338743998001783,
      "language_loss": 0.66352236,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.68551838,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.85756778717041
    },
    {
      "auxiliary_loss_clip": 0.01189154,
      "auxiliary_loss_mlp": 0.01031574,
      "balance_loss_clip": 1.05607486,
      "balance_loss_mlp": 1.02318752,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 1.8728189938297224,
      "language_loss": 0.80266321,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.82487047,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 2.6038951873779297
    },
    {
      "auxiliary_loss_clip": 0.01172912,
      "auxiliary_loss_mlp": 0.01029202,
      "balance_loss_clip": 0.97294188,
      "balance_loss_mlp": 1.02089322,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 1.8967904167965146,
      "language_loss": 0.81934106,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.84136224,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.7099010944366455
    },
    {
      "auxiliary_loss_clip": 0.01184402,
      "auxiliary_loss_mlp": 0.01025813,
      "balance_loss_clip": 1.0527029,
      "balance_loss_mlp": 1.01811159,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 1.604825398059585,
      "language_loss": 0.82452309,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.84662521,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 3.6980745792388916
    },
    {
      "auxiliary_loss_clip": 0.01171922,
      "auxiliary_loss_mlp": 0.01034045,
      "balance_loss_clip": 0.97352666,
      "balance_loss_mlp": 1.02593899,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 2.8639397735687173,
      "language_loss": 0.82087505,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.84293473,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 2.65509033203125
    },
    {
      "auxiliary_loss_clip": 0.01182281,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 0.93727344,
      "balance_loss_mlp": 1.01867402,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 2.096468028140605,
      "language_loss": 0.90096587,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.92305779,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 3.646451950073242
    },
    {
      "auxiliary_loss_clip": 0.01186903,
      "auxiliary_loss_mlp": 0.01031709,
      "balance_loss_clip": 1.0533936,
      "balance_loss_mlp": 1.02332854,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.1924866330246613,
      "language_loss": 0.81287098,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83505708,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.731510877609253
    },
    {
      "auxiliary_loss_clip": 0.0118621,
      "auxiliary_loss_mlp": 0.01027489,
      "balance_loss_clip": 1.01604533,
      "balance_loss_mlp": 1.01935339,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 2.4450000178683453,
      "language_loss": 0.64261198,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66474903,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.7443079948425293
    },
    {
      "auxiliary_loss_clip": 0.01181522,
      "auxiliary_loss_mlp": 0.01030405,
      "balance_loss_clip": 1.01226115,
      "balance_loss_mlp": 1.02254915,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 1.8190209877442687,
      "language_loss": 0.6031642,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62528348,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.7650699615478516
    },
    {
      "auxiliary_loss_clip": 0.01174816,
      "auxiliary_loss_mlp": 0.0103265,
      "balance_loss_clip": 0.93373203,
      "balance_loss_mlp": 1.02407885,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 1.7581176877423315,
      "language_loss": 0.76048255,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.78255725,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.7725958824157715
    },
    {
      "auxiliary_loss_clip": 0.01185268,
      "auxiliary_loss_mlp": 0.01030009,
      "balance_loss_clip": 1.01642704,
      "balance_loss_mlp": 1.02209949,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.6366598316776273,
      "language_loss": 0.70920825,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.73136103,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 3.5526251792907715
    },
    {
      "auxiliary_loss_clip": 0.01090252,
      "auxiliary_loss_mlp": 0.01004046,
      "balance_loss_clip": 0.9088217,
      "balance_loss_mlp": 1.00181723,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.8409357208019408,
      "language_loss": 0.59541738,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61636043,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.4727022647857666
    },
    {
      "auxiliary_loss_clip": 0.0118206,
      "auxiliary_loss_mlp": 0.01035033,
      "balance_loss_clip": 1.0119704,
      "balance_loss_mlp": 1.02716517,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 2.081497345402461,
      "language_loss": 0.79841208,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.82058299,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.848552703857422
    },
    {
      "auxiliary_loss_clip": 0.01181138,
      "auxiliary_loss_mlp": 0.01033957,
      "balance_loss_clip": 0.94115716,
      "balance_loss_mlp": 1.02642894,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 2.1052318539259014,
      "language_loss": 0.74542338,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.76757431,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 3.5338706970214844
    },
    {
      "auxiliary_loss_clip": 0.01170893,
      "auxiliary_loss_mlp": 0.01025209,
      "balance_loss_clip": 0.89860392,
      "balance_loss_mlp": 1.01784766,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.4810956732422744,
      "language_loss": 0.68378627,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70574731,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.7484281063079834
    },
    {
      "auxiliary_loss_clip": 0.0118335,
      "auxiliary_loss_mlp": 0.01024485,
      "balance_loss_clip": 1.01807189,
      "balance_loss_mlp": 1.01722848,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 1.7467121686506693,
      "language_loss": 0.71058673,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73266506,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.5602097511291504
    },
    {
      "auxiliary_loss_clip": 0.01184225,
      "auxiliary_loss_mlp": 0.01034819,
      "balance_loss_clip": 0.93635857,
      "balance_loss_mlp": 1.02594972,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 2.299199851944006,
      "language_loss": 0.83448946,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.85667986,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 2.5673296451568604
    },
    {
      "auxiliary_loss_clip": 0.01190586,
      "auxiliary_loss_mlp": 0.01124484,
      "balance_loss_clip": 1.02029085,
      "balance_loss_mlp": 0.0,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.2266061132696553,
      "language_loss": 0.64628965,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.66944033,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.5867016315460205
    },
    {
      "auxiliary_loss_clip": 0.01184149,
      "auxiliary_loss_mlp": 0.01032435,
      "balance_loss_clip": 1.05385518,
      "balance_loss_mlp": 1.02434087,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 1.9431797368239792,
      "language_loss": 0.72700131,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.74916714,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.7353382110595703
    },
    {
      "auxiliary_loss_clip": 0.01183145,
      "auxiliary_loss_mlp": 0.01029762,
      "balance_loss_clip": 0.94132084,
      "balance_loss_mlp": 1.02217412,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 2.420702232201793,
      "language_loss": 0.77540088,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.79752994,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.6192286014556885
    },
    {
      "auxiliary_loss_clip": 0.01176392,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.01347852,
      "balance_loss_mlp": 1.01621175,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 1.778526670227745,
      "language_loss": 0.79086423,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81287616,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 2.750223159790039
    },
    {
      "auxiliary_loss_clip": 0.01179326,
      "auxiliary_loss_mlp": 0.01031943,
      "balance_loss_clip": 0.90099061,
      "balance_loss_mlp": 1.02379823,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 1.6816348748816121,
      "language_loss": 0.70369691,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.72580963,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.741786479949951
    },
    {
      "auxiliary_loss_clip": 0.01185127,
      "auxiliary_loss_mlp": 0.01032618,
      "balance_loss_clip": 1.05432498,
      "balance_loss_mlp": 1.02500057,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 2.0999580816504495,
      "language_loss": 0.76263797,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78481537,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 2.646040678024292
    },
    {
      "auxiliary_loss_clip": 0.01186081,
      "auxiliary_loss_mlp": 0.01032848,
      "balance_loss_clip": 1.05451143,
      "balance_loss_mlp": 1.02473009,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 1.7235008665520934,
      "language_loss": 0.78457916,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.80676842,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 2.6106810569763184
    },
    {
      "auxiliary_loss_clip": 0.01172654,
      "auxiliary_loss_mlp": 0.01030357,
      "balance_loss_clip": 0.89549804,
      "balance_loss_mlp": 1.0215292,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.3099794341627637,
      "language_loss": 0.82846749,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.8504976,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 2.6980061531066895
    },
    {
      "auxiliary_loss_clip": 0.01175641,
      "auxiliary_loss_mlp": 0.01029982,
      "balance_loss_clip": 1.01153064,
      "balance_loss_mlp": 1.02187562,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 1.8981760245023633,
      "language_loss": 0.85354257,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87559891,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 2.6573944091796875
    },
    {
      "auxiliary_loss_clip": 0.01182085,
      "auxiliary_loss_mlp": 0.01028825,
      "balance_loss_clip": 0.97608936,
      "balance_loss_mlp": 1.02116275,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 1.8274751943503678,
      "language_loss": 0.7110545,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.7331636,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 2.791857957839966
    },
    {
      "auxiliary_loss_clip": 0.01185978,
      "auxiliary_loss_mlp": 0.01029157,
      "balance_loss_clip": 1.01130342,
      "balance_loss_mlp": 1.02066946,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 1.6862101041437152,
      "language_loss": 0.76842684,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79057819,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.832670211791992
    },
    {
      "auxiliary_loss_clip": 0.0117852,
      "auxiliary_loss_mlp": 0.01020961,
      "balance_loss_clip": 1.01437044,
      "balance_loss_mlp": 1.01365948,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.5929471238328088,
      "language_loss": 0.81149465,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83348948,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.7578775882720947
    },
    {
      "auxiliary_loss_clip": 0.01185438,
      "auxiliary_loss_mlp": 0.01036274,
      "balance_loss_clip": 1.01536977,
      "balance_loss_mlp": 1.02805424,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 1.7455175036804593,
      "language_loss": 0.81033683,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.83255392,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.679290771484375
    },
    {
      "auxiliary_loss_clip": 0.01190471,
      "auxiliary_loss_mlp": 0.01032692,
      "balance_loss_clip": 1.01804852,
      "balance_loss_mlp": 1.02474713,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 2.143308680898854,
      "language_loss": 0.87359387,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89582551,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 3.610063314437866
    },
    {
      "auxiliary_loss_clip": 0.0117443,
      "auxiliary_loss_mlp": 0.01032148,
      "balance_loss_clip": 1.01248443,
      "balance_loss_mlp": 1.02430987,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.5908534833493024,
      "language_loss": 0.79073751,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81280327,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 3.6309571266174316
    },
    {
      "auxiliary_loss_clip": 0.01175741,
      "auxiliary_loss_mlp": 0.01031192,
      "balance_loss_clip": 0.97341001,
      "balance_loss_mlp": 1.0234375,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.4883849080597344,
      "language_loss": 0.8245821,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.84665143,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.776477575302124
    },
    {
      "auxiliary_loss_clip": 0.01186597,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 0.94105685,
      "balance_loss_mlp": 1.02138543,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.6214684815419076,
      "language_loss": 0.7646848,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.78684425,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.8066282272338867
    },
    {
      "auxiliary_loss_clip": 0.01190108,
      "auxiliary_loss_mlp": 0.01033168,
      "balance_loss_clip": 1.01551771,
      "balance_loss_mlp": 1.02492428,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 2.0711892419832925,
      "language_loss": 0.89779115,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92002392,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.7094619274139404
    },
    {
      "auxiliary_loss_clip": 0.01184051,
      "auxiliary_loss_mlp": 0.01027207,
      "balance_loss_clip": 0.97785324,
      "balance_loss_mlp": 1.01953042,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 1.4981002109707955,
      "language_loss": 0.76823837,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.79035097,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.7265350818634033
    },
    {
      "auxiliary_loss_clip": 0.01183981,
      "auxiliary_loss_mlp": 0.01033466,
      "balance_loss_clip": 0.9739567,
      "balance_loss_mlp": 1.02503181,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 1.891617404389135,
      "language_loss": 0.81730866,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.83948314,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 3.7521355152130127
    },
    {
      "auxiliary_loss_clip": 0.01182112,
      "auxiliary_loss_mlp": 0.01030483,
      "balance_loss_clip": 1.01679468,
      "balance_loss_mlp": 1.02283299,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.924627931081606,
      "language_loss": 0.66628385,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68840981,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.692183494567871
    },
    {
      "auxiliary_loss_clip": 0.01188353,
      "auxiliary_loss_mlp": 0.01035502,
      "balance_loss_clip": 1.01416993,
      "balance_loss_mlp": 1.0258882,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 2.059131853907219,
      "language_loss": 0.71814942,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.74038798,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.84641432762146
    },
    {
      "auxiliary_loss_clip": 0.01181534,
      "auxiliary_loss_mlp": 0.01027805,
      "balance_loss_clip": 0.93837863,
      "balance_loss_mlp": 1.02014589,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 2.16578060458909,
      "language_loss": 0.88584656,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.90793997,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 3.583651304244995
    },
    {
      "auxiliary_loss_clip": 0.01083591,
      "auxiliary_loss_mlp": 0.0100216,
      "balance_loss_clip": 0.94369394,
      "balance_loss_mlp": 0.99978751,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.7592595534061417,
      "language_loss": 0.4862465,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50710398,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.359290838241577
    },
    {
      "auxiliary_loss_clip": 0.01184722,
      "auxiliary_loss_mlp": 0.01027722,
      "balance_loss_clip": 0.90419596,
      "balance_loss_mlp": 1.01944304,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 1.840865773063439,
      "language_loss": 0.89709109,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.91921556,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.7780370712280273
    },
    {
      "auxiliary_loss_clip": 0.01187284,
      "auxiliary_loss_mlp": 0.01123803,
      "balance_loss_clip": 1.01638138,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 1.6650443890055666,
      "language_loss": 0.86924887,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.89235973,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.76521635055542
    },
    {
      "auxiliary_loss_clip": 0.01180922,
      "auxiliary_loss_mlp": 0.01030079,
      "balance_loss_clip": 0.9772315,
      "balance_loss_mlp": 1.02196074,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.1049631309111794,
      "language_loss": 0.65917826,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.6812883,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 2.785198450088501
    },
    {
      "auxiliary_loss_clip": 0.01182827,
      "auxiliary_loss_mlp": 0.01031164,
      "balance_loss_clip": 1.05348384,
      "balance_loss_mlp": 1.02336216,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 3.018501101515092,
      "language_loss": 0.80846536,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.83060527,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.7450766563415527
    },
    {
      "auxiliary_loss_clip": 0.01078465,
      "auxiliary_loss_mlp": 0.00999253,
      "balance_loss_clip": 0.98113751,
      "balance_loss_mlp": 0.99714327,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7725313996892227,
      "language_loss": 0.60539126,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62616849,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.335956573486328
    },
    {
      "auxiliary_loss_clip": 0.01181728,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.01350427,
      "balance_loss_mlp": 1.0207324,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 1.7119847576433478,
      "language_loss": 0.68023109,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70233917,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 2.7528903484344482
    },
    {
      "auxiliary_loss_clip": 0.01179412,
      "auxiliary_loss_mlp": 0.0102481,
      "balance_loss_clip": 0.97629803,
      "balance_loss_mlp": 1.01680541,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 2.041522861496513,
      "language_loss": 0.81424022,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.83628243,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 2.8873562812805176
    },
    {
      "auxiliary_loss_clip": 0.01189417,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.01605356,
      "balance_loss_mlp": 1.01712155,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 1.9518455918513153,
      "language_loss": 0.83295786,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85510498,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 2.692791700363159
    },
    {
      "auxiliary_loss_clip": 0.01189123,
      "auxiliary_loss_mlp": 0.01028573,
      "balance_loss_clip": 1.01686573,
      "balance_loss_mlp": 1.02044845,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 1.4772941939043687,
      "language_loss": 0.84631026,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.86848724,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.7546722888946533
    },
    {
      "auxiliary_loss_clip": 0.01183982,
      "auxiliary_loss_mlp": 0.01030612,
      "balance_loss_clip": 0.9403975,
      "balance_loss_mlp": 1.02236915,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 1.885813981351007,
      "language_loss": 0.62118673,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64333266,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 2.761237621307373
    },
    {
      "auxiliary_loss_clip": 0.01168213,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 0.93504393,
      "balance_loss_mlp": 1.01930761,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.4902215024663361,
      "language_loss": 0.83900076,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.86095321,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 2.745850086212158
    },
    {
      "auxiliary_loss_clip": 0.01185038,
      "auxiliary_loss_mlp": 0.01027641,
      "balance_loss_clip": 1.01401794,
      "balance_loss_mlp": 1.01983225,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 4.860791257553615,
      "language_loss": 0.75045037,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.77257717,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 2.6638104915618896
    },
    {
      "auxiliary_loss_clip": 0.0117983,
      "auxiliary_loss_mlp": 0.01032343,
      "balance_loss_clip": 1.01535428,
      "balance_loss_mlp": 1.0246954,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 1.331093894591875,
      "language_loss": 0.875691,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.89781272,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.803112030029297
    },
    {
      "auxiliary_loss_clip": 0.01188863,
      "auxiliary_loss_mlp": 0.01033428,
      "balance_loss_clip": 0.97802782,
      "balance_loss_mlp": 1.02492273,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 1.9544506582213945,
      "language_loss": 0.76206207,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78428501,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.701558828353882
    },
    {
      "auxiliary_loss_clip": 0.01189789,
      "auxiliary_loss_mlp": 0.01030119,
      "balance_loss_clip": 1.01676464,
      "balance_loss_mlp": 1.02176249,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 2.3030150733336914,
      "language_loss": 0.79269201,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.81489116,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 2.6888787746429443
    },
    {
      "auxiliary_loss_clip": 0.01190714,
      "auxiliary_loss_mlp": 0.01031216,
      "balance_loss_clip": 0.97738528,
      "balance_loss_mlp": 1.02258539,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 1.983527064487312,
      "language_loss": 0.82306749,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84528685,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 3.6531152725219727
    },
    {
      "auxiliary_loss_clip": 0.01179157,
      "auxiliary_loss_mlp": 0.01029011,
      "balance_loss_clip": 0.93752337,
      "balance_loss_mlp": 1.02135801,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 1.9444293587599286,
      "language_loss": 0.84869313,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.87077487,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 3.638740301132202
    },
    {
      "auxiliary_loss_clip": 0.01184604,
      "auxiliary_loss_mlp": 0.01037562,
      "balance_loss_clip": 0.90564835,
      "balance_loss_mlp": 1.0292654,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 2.064398562076332,
      "language_loss": 0.74479401,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.7670157,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.820843458175659
    },
    {
      "auxiliary_loss_clip": 0.01184445,
      "auxiliary_loss_mlp": 0.01029878,
      "balance_loss_clip": 1.01512885,
      "balance_loss_mlp": 1.02093124,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 2.0360174892908804,
      "language_loss": 0.8202424,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84238565,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.646605968475342
    },
    {
      "auxiliary_loss_clip": 0.01172378,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 0.93408358,
      "balance_loss_mlp": 1.01775348,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 6.890380291113251,
      "language_loss": 0.75592583,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77791393,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.782953977584839
    },
    {
      "auxiliary_loss_clip": 0.01182261,
      "auxiliary_loss_mlp": 0.01034995,
      "balance_loss_clip": 0.97326452,
      "balance_loss_mlp": 1.02649879,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 1.9304803207013959,
      "language_loss": 0.92538971,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.94756234,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.747084856033325
    },
    {
      "auxiliary_loss_clip": 0.01189979,
      "auxiliary_loss_mlp": 0.01030871,
      "balance_loss_clip": 1.01388884,
      "balance_loss_mlp": 1.02287221,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 2.778310525542392,
      "language_loss": 0.90938562,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.93159413,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 3.8118298053741455
    },
    {
      "auxiliary_loss_clip": 0.01173591,
      "auxiliary_loss_mlp": 0.01032475,
      "balance_loss_clip": 1.01086009,
      "balance_loss_mlp": 1.02432704,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 1.5772704615264874,
      "language_loss": 0.67661303,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.69867373,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.7366650104522705
    },
    {
      "auxiliary_loss_clip": 0.01184648,
      "auxiliary_loss_mlp": 0.01031553,
      "balance_loss_clip": 1.01457489,
      "balance_loss_mlp": 1.0232743,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 2.207988280705494,
      "language_loss": 0.77201593,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.79417801,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 3.6512107849121094
    },
    {
      "auxiliary_loss_clip": 0.01173243,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 0.97550392,
      "balance_loss_mlp": 1.02064681,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.7156138059156698,
      "language_loss": 0.72203344,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74405378,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.805002212524414
    },
    {
      "auxiliary_loss_clip": 0.01190611,
      "auxiliary_loss_mlp": 0.0103164,
      "balance_loss_clip": 0.97856975,
      "balance_loss_mlp": 1.02352822,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 2.555219731340438,
      "language_loss": 0.8331607,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.85538322,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.714872360229492
    },
    {
      "auxiliary_loss_clip": 0.01185535,
      "auxiliary_loss_mlp": 0.01029199,
      "balance_loss_clip": 1.05336809,
      "balance_loss_mlp": 1.02161765,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 2.2672093910541657,
      "language_loss": 0.76353168,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.78567898,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.5987908840179443
    },
    {
      "auxiliary_loss_clip": 0.01181584,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 0.9380272,
      "balance_loss_mlp": 1.02315974,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.741333351568676,
      "language_loss": 0.66618681,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.68830949,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.8513870239257812
    },
    {
      "auxiliary_loss_clip": 0.01169792,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 0.97211981,
      "balance_loss_mlp": 1.01891398,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.6146545494157338,
      "language_loss": 0.65801257,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.67998087,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 2.7905654907226562
    },
    {
      "auxiliary_loss_clip": 0.01183624,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.0539844,
      "balance_loss_mlp": 1.02259612,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 1.4263117820541922,
      "language_loss": 0.70432627,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.72646785,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.702061891555786
    },
    {
      "auxiliary_loss_clip": 0.01181245,
      "auxiliary_loss_mlp": 0.01028733,
      "balance_loss_clip": 1.01353121,
      "balance_loss_mlp": 1.02075195,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.2358879872527804,
      "language_loss": 0.76342523,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.78552508,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.646394729614258
    },
    {
      "auxiliary_loss_clip": 0.01178664,
      "auxiliary_loss_mlp": 0.01031607,
      "balance_loss_clip": 0.98040438,
      "balance_loss_mlp": 1.02394795,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 2.010147805463774,
      "language_loss": 0.79514122,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.81724393,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 2.7372541427612305
    },
    {
      "auxiliary_loss_clip": 0.01186535,
      "auxiliary_loss_mlp": 0.01124232,
      "balance_loss_clip": 1.01417351,
      "balance_loss_mlp": 0.0,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 2.388234275116468,
      "language_loss": 0.7720539,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79516155,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 2.698164463043213
    },
    {
      "auxiliary_loss_clip": 0.01187175,
      "auxiliary_loss_mlp": 0.0103286,
      "balance_loss_clip": 1.0562892,
      "balance_loss_mlp": 1.02492642,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.2311707881388383,
      "language_loss": 0.87815326,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.90035355,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.625833749771118
    },
    {
      "auxiliary_loss_clip": 0.01184585,
      "auxiliary_loss_mlp": 0.0103295,
      "balance_loss_clip": 1.01605439,
      "balance_loss_mlp": 1.02520752,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 2.3333621862068865,
      "language_loss": 0.74483716,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.76701248,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 2.7949509620666504
    },
    {
      "auxiliary_loss_clip": 0.0118663,
      "auxiliary_loss_mlp": 0.01028665,
      "balance_loss_clip": 1.01781416,
      "balance_loss_mlp": 1.02085042,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 1.7554700761502762,
      "language_loss": 0.82800448,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.8501575,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 2.857043743133545
    },
    {
      "auxiliary_loss_clip": 0.01188637,
      "auxiliary_loss_mlp": 0.01027332,
      "balance_loss_clip": 1.01600671,
      "balance_loss_mlp": 1.01930308,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.1553183481869707,
      "language_loss": 0.70802808,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.73018777,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 2.7343413829803467
    },
    {
      "auxiliary_loss_clip": 0.01178985,
      "auxiliary_loss_mlp": 0.0103367,
      "balance_loss_clip": 1.01339912,
      "balance_loss_mlp": 1.02553988,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 1.7095404372921565,
      "language_loss": 0.93562919,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.95775568,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 2.654963493347168
    },
    {
      "auxiliary_loss_clip": 0.01077604,
      "auxiliary_loss_mlp": 0.01003372,
      "balance_loss_clip": 0.98064816,
      "balance_loss_mlp": 1.00132179,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.864835512376622,
      "language_loss": 0.59848249,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61929226,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.0938093662261963
    },
    {
      "auxiliary_loss_clip": 0.01185555,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 0.97796524,
      "balance_loss_mlp": 1.024791,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 3.4231892871103793,
      "language_loss": 0.72819149,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.75038189,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.667785167694092
    },
    {
      "auxiliary_loss_clip": 0.01186046,
      "auxiliary_loss_mlp": 0.01029518,
      "balance_loss_clip": 1.0553093,
      "balance_loss_mlp": 1.02186728,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 1.7053465460060788,
      "language_loss": 0.67799968,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70015538,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 3.63136887550354
    },
    {
      "auxiliary_loss_clip": 0.01177291,
      "auxiliary_loss_mlp": 0.01027151,
      "balance_loss_clip": 0.89710116,
      "balance_loss_mlp": 1.01882422,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 3.4259126118155656,
      "language_loss": 0.81991696,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.84196138,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 2.748756170272827
    },
    {
      "auxiliary_loss_clip": 0.01073143,
      "auxiliary_loss_mlp": 0.01001389,
      "balance_loss_clip": 0.90516001,
      "balance_loss_mlp": 0.9992556,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.8410169601019017,
      "language_loss": 0.57375276,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59449816,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 3.9894216060638428
    },
    {
      "auxiliary_loss_clip": 0.01185898,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 0.97917175,
      "balance_loss_mlp": 1.01988959,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.7183353726087245,
      "language_loss": 0.77256596,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.79470199,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.674785852432251
    },
    {
      "auxiliary_loss_clip": 0.01189989,
      "auxiliary_loss_mlp": 0.01031832,
      "balance_loss_clip": 1.01813996,
      "balance_loss_mlp": 1.02385664,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 15.901095416584202,
      "language_loss": 0.74034154,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.76255977,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.547853469848633
    },
    {
      "auxiliary_loss_clip": 0.01185683,
      "auxiliary_loss_mlp": 0.01030101,
      "balance_loss_clip": 1.01864636,
      "balance_loss_mlp": 1.02196455,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.8013195837929044,
      "language_loss": 0.85287666,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87503451,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.6352100372314453
    },
    {
      "auxiliary_loss_clip": 0.01170442,
      "auxiliary_loss_mlp": 0.01028543,
      "balance_loss_clip": 0.89945006,
      "balance_loss_mlp": 1.01992416,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 2.1090235126065693,
      "language_loss": 0.77693933,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.79892921,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.8280868530273438
    },
    {
      "auxiliary_loss_clip": 0.01183655,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 0.97690713,
      "balance_loss_mlp": 1.02595472,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 2.328682146898751,
      "language_loss": 0.78519917,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.80737495,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 3.7552907466888428
    },
    {
      "auxiliary_loss_clip": 0.0116417,
      "auxiliary_loss_mlp": 0.01033061,
      "balance_loss_clip": 0.9336791,
      "balance_loss_mlp": 1.02486491,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.0365625849491504,
      "language_loss": 0.80864441,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.83061671,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.721510171890259
    },
    {
      "auxiliary_loss_clip": 0.01184599,
      "auxiliary_loss_mlp": 0.01123267,
      "balance_loss_clip": 1.01798809,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.5396730148816795,
      "language_loss": 0.85482216,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.87790084,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 3.6302592754364014
    },
    {
      "auxiliary_loss_clip": 0.01188095,
      "auxiliary_loss_mlp": 0.01028607,
      "balance_loss_clip": 1.05470908,
      "balance_loss_mlp": 1.01955354,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 1.5312920615128378,
      "language_loss": 0.75772119,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.77988827,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.7237744331359863
    },
    {
      "auxiliary_loss_clip": 0.01182743,
      "auxiliary_loss_mlp": 0.01031648,
      "balance_loss_clip": 0.9767822,
      "balance_loss_mlp": 1.02282679,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 2.1384718138942604,
      "language_loss": 0.87376595,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89590979,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.6532421112060547
    },
    {
      "auxiliary_loss_clip": 0.01186862,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.01672196,
      "balance_loss_mlp": 1.01812327,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 4.177629031255262,
      "language_loss": 0.68538666,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.70751905,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.63486647605896
    },
    {
      "auxiliary_loss_clip": 0.01181111,
      "auxiliary_loss_mlp": 0.01030065,
      "balance_loss_clip": 1.01684678,
      "balance_loss_mlp": 1.02228689,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 1.756850079218706,
      "language_loss": 0.71510178,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.73721361,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.6636526584625244
    },
    {
      "auxiliary_loss_clip": 0.01176271,
      "auxiliary_loss_mlp": 0.01035867,
      "balance_loss_clip": 0.82345021,
      "balance_loss_mlp": 1.02699733,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 1.9663959702263367,
      "language_loss": 0.69974673,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.7218681,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.8546142578125
    },
    {
      "auxiliary_loss_clip": 0.01180427,
      "auxiliary_loss_mlp": 0.01027958,
      "balance_loss_clip": 0.98047304,
      "balance_loss_mlp": 1.02045345,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 1.5042982834302254,
      "language_loss": 0.82486373,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84694761,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 2.866251230239868
    },
    {
      "auxiliary_loss_clip": 0.01181765,
      "auxiliary_loss_mlp": 0.0103409,
      "balance_loss_clip": 0.97677493,
      "balance_loss_mlp": 1.02587044,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 1.8764145419878338,
      "language_loss": 0.72370481,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74586332,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 2.726776599884033
    },
    {
      "auxiliary_loss_clip": 0.01187799,
      "auxiliary_loss_mlp": 0.01034847,
      "balance_loss_clip": 0.97969306,
      "balance_loss_mlp": 1.02662122,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 2.599894586333351,
      "language_loss": 0.84317571,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86540216,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.7294960021972656
    },
    {
      "auxiliary_loss_clip": 0.01174074,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 0.97012758,
      "balance_loss_mlp": 1.01916778,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 1.7730565647167738,
      "language_loss": 0.8292948,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85129905,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 2.6541099548339844
    },
    {
      "auxiliary_loss_clip": 0.01181918,
      "auxiliary_loss_mlp": 0.01028109,
      "balance_loss_clip": 1.01397359,
      "balance_loss_mlp": 1.02021098,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 1.8841829443617668,
      "language_loss": 0.7387076,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76080787,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 2.664508581161499
    },
    {
      "auxiliary_loss_clip": 0.01183634,
      "auxiliary_loss_mlp": 0.01031829,
      "balance_loss_clip": 1.01429534,
      "balance_loss_mlp": 1.02353823,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 1.7906105046929097,
      "language_loss": 0.7905857,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.81274033,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.6346659660339355
    },
    {
      "auxiliary_loss_clip": 0.01192919,
      "auxiliary_loss_mlp": 0.01028047,
      "balance_loss_clip": 1.01657701,
      "balance_loss_mlp": 1.01924372,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 2.481663643288022,
      "language_loss": 0.80168283,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.82389247,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 2.688317060470581
    },
    {
      "auxiliary_loss_clip": 0.01186833,
      "auxiliary_loss_mlp": 0.01027264,
      "balance_loss_clip": 0.9785198,
      "balance_loss_mlp": 1.01921153,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 2.264699992972252,
      "language_loss": 0.72369158,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.74583256,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.7390100955963135
    },
    {
      "auxiliary_loss_clip": 0.01183631,
      "auxiliary_loss_mlp": 0.01028957,
      "balance_loss_clip": 0.89932102,
      "balance_loss_mlp": 1.02109492,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 1.490910499937179,
      "language_loss": 0.78925937,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.81138527,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 2.8695313930511475
    },
    {
      "auxiliary_loss_clip": 0.0118466,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.0547241,
      "balance_loss_mlp": 1.02145147,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 1.8840471790597813,
      "language_loss": 0.78525078,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.80739009,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 2.6245949268341064
    },
    {
      "auxiliary_loss_clip": 0.01185185,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.01621735,
      "balance_loss_mlp": 1.01798236,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 1.7262719566726883,
      "language_loss": 0.88804317,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.91015768,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 2.6534059047698975
    },
    {
      "auxiliary_loss_clip": 0.01177247,
      "auxiliary_loss_mlp": 0.01023365,
      "balance_loss_clip": 0.97790134,
      "balance_loss_mlp": 1.01559234,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.8203920743258069,
      "language_loss": 0.86267853,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88468468,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.705723762512207
    },
    {
      "auxiliary_loss_clip": 0.01177376,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 0.89473766,
      "balance_loss_mlp": 1.01890814,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.6661504436160866,
      "language_loss": 0.8111366,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.83318138,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 4.021667003631592
    },
    {
      "auxiliary_loss_clip": 0.01185143,
      "auxiliary_loss_mlp": 0.0103456,
      "balance_loss_clip": 1.0530107,
      "balance_loss_mlp": 1.02627516,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 1.7405422447748349,
      "language_loss": 0.69377917,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.71597624,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 3.6665024757385254
    },
    {
      "auxiliary_loss_clip": 0.01186912,
      "auxiliary_loss_mlp": 0.01025693,
      "balance_loss_clip": 0.97630316,
      "balance_loss_mlp": 1.01802826,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.0238108103088988,
      "language_loss": 0.8155483,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.83767432,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.7223217487335205
    },
    {
      "auxiliary_loss_clip": 0.01090144,
      "auxiliary_loss_mlp": 0.01007737,
      "balance_loss_clip": 0.90540528,
      "balance_loss_mlp": 1.00569844,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.7768083028587515,
      "language_loss": 0.56884819,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.589827,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.340073347091675
    },
    {
      "auxiliary_loss_clip": 0.01188664,
      "auxiliary_loss_mlp": 0.0112381,
      "balance_loss_clip": 1.01739526,
      "balance_loss_mlp": 0.0,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 1.7909302879774556,
      "language_loss": 0.7173506,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.7404753,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.714311122894287
    },
    {
      "auxiliary_loss_clip": 0.01179261,
      "auxiliary_loss_mlp": 0.01027523,
      "balance_loss_clip": 1.01195478,
      "balance_loss_mlp": 1.01989377,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.6709588102739086,
      "language_loss": 0.7284168,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75048459,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.6547048091888428
    },
    {
      "auxiliary_loss_clip": 0.0118167,
      "auxiliary_loss_mlp": 0.01028159,
      "balance_loss_clip": 0.93730116,
      "balance_loss_mlp": 1.02017784,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 1.5982720706408504,
      "language_loss": 0.89910173,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92120004,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 3.7038562297821045
    },
    {
      "auxiliary_loss_clip": 0.01188944,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 0.93869668,
      "balance_loss_mlp": 1.01956177,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 2.3588163445025634,
      "language_loss": 0.78642607,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.80858767,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.6946094036102295
    },
    {
      "auxiliary_loss_clip": 0.01184895,
      "auxiliary_loss_mlp": 0.01032481,
      "balance_loss_clip": 0.93720877,
      "balance_loss_mlp": 1.02475619,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 1.7071027898128648,
      "language_loss": 0.73694044,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.75911427,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.6801202297210693
    },
    {
      "auxiliary_loss_clip": 0.01172762,
      "auxiliary_loss_mlp": 0.01027699,
      "balance_loss_clip": 0.93727338,
      "balance_loss_mlp": 1.01931858,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 1.58524868461853,
      "language_loss": 0.75147271,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77347738,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 3.6729164123535156
    },
    {
      "auxiliary_loss_clip": 0.01182714,
      "auxiliary_loss_mlp": 0.01022518,
      "balance_loss_clip": 1.01708722,
      "balance_loss_mlp": 1.01471007,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 1.6614025902319216,
      "language_loss": 0.72669643,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.74874878,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.7098143100738525
    },
    {
      "auxiliary_loss_clip": 0.01183224,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 1.05178308,
      "balance_loss_mlp": 1.01907182,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 1.771713530558185,
      "language_loss": 0.79574078,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81784534,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.66764235496521
    },
    {
      "auxiliary_loss_clip": 0.01079442,
      "auxiliary_loss_mlp": 0.01007052,
      "balance_loss_clip": 1.01767957,
      "balance_loss_mlp": 1.00500154,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.8903898146061476,
      "language_loss": 0.57002759,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.59089249,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.2156879901885986
    },
    {
      "auxiliary_loss_clip": 0.01177251,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.01317191,
      "balance_loss_mlp": 1.02287471,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 1.9086926314623047,
      "language_loss": 0.69528961,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.7173667,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.798804759979248
    },
    {
      "auxiliary_loss_clip": 0.01174681,
      "auxiliary_loss_mlp": 0.01123876,
      "balance_loss_clip": 0.97432959,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 1.9951859303044275,
      "language_loss": 0.75332314,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.77630872,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.805312156677246
    },
    {
      "auxiliary_loss_clip": 0.01175328,
      "auxiliary_loss_mlp": 0.01026083,
      "balance_loss_clip": 0.93513429,
      "balance_loss_mlp": 1.01800692,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.7101309530706774,
      "language_loss": 0.85073757,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87275171,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 2.725778818130493
    },
    {
      "auxiliary_loss_clip": 0.01182048,
      "auxiliary_loss_mlp": 0.01026553,
      "balance_loss_clip": 0.93711126,
      "balance_loss_mlp": 1.01889932,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 1.9381004556094092,
      "language_loss": 0.78743261,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.80951864,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.7486889362335205
    },
    {
      "auxiliary_loss_clip": 0.01187826,
      "auxiliary_loss_mlp": 0.01026475,
      "balance_loss_clip": 0.89895982,
      "balance_loss_mlp": 1.01867843,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 1.945686416946306,
      "language_loss": 0.78775311,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.80989611,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.719949245452881
    },
    {
      "auxiliary_loss_clip": 0.01185452,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 1.05383217,
      "balance_loss_mlp": 1.02114487,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 1.8839705079653524,
      "language_loss": 0.84218788,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.86432832,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 2.6288106441497803
    },
    {
      "auxiliary_loss_clip": 0.011869,
      "auxiliary_loss_mlp": 0.01034079,
      "balance_loss_clip": 1.0527482,
      "balance_loss_mlp": 1.02580643,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 1.8915413824717922,
      "language_loss": 0.84491003,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.86711979,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 2.646559000015259
    },
    {
      "auxiliary_loss_clip": 0.01180858,
      "auxiliary_loss_mlp": 0.01124612,
      "balance_loss_clip": 0.97721398,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.699573364033841,
      "language_loss": 0.81295574,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.83601046,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 2.7384495735168457
    },
    {
      "auxiliary_loss_clip": 0.01091856,
      "auxiliary_loss_mlp": 0.01004832,
      "balance_loss_clip": 0.90705454,
      "balance_loss_mlp": 1.00267398,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.8580401482032136,
      "language_loss": 0.60918498,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.63015187,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.4149351119995117
    },
    {
      "auxiliary_loss_clip": 0.01160312,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 0.89549148,
      "balance_loss_mlp": 1.02345347,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 1.8308780932224726,
      "language_loss": 0.79184508,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.81376207,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 3.0254228115081787
    },
    {
      "auxiliary_loss_clip": 0.0116705,
      "auxiliary_loss_mlp": 0.01028361,
      "balance_loss_clip": 0.97230744,
      "balance_loss_mlp": 1.02069557,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 1.7984585529942891,
      "language_loss": 0.84878957,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.87074363,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 2.7845592498779297
    },
    {
      "auxiliary_loss_clip": 0.01175904,
      "auxiliary_loss_mlp": 0.01029619,
      "balance_loss_clip": 1.01008272,
      "balance_loss_mlp": 1.02204275,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 1.814998442435533,
      "language_loss": 0.84617609,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.8682313,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.6845178604125977
    },
    {
      "auxiliary_loss_clip": 0.01186695,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 0.97966409,
      "balance_loss_mlp": 1.02266359,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.0303442890777306,
      "language_loss": 0.83937621,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86155713,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 2.652322292327881
    },
    {
      "auxiliary_loss_clip": 0.01184474,
      "auxiliary_loss_mlp": 0.01024609,
      "balance_loss_clip": 1.05368078,
      "balance_loss_mlp": 1.01643097,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.443513959197461,
      "language_loss": 0.78599334,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80808425,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 3.6564278602600098
    },
    {
      "auxiliary_loss_clip": 0.01183337,
      "auxiliary_loss_mlp": 0.01030108,
      "balance_loss_clip": 1.01559067,
      "balance_loss_mlp": 1.02164376,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 2.3694012889912357,
      "language_loss": 0.73836988,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.76050436,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 3.588388681411743
    },
    {
      "auxiliary_loss_clip": 0.01175429,
      "auxiliary_loss_mlp": 0.01024607,
      "balance_loss_clip": 0.85782599,
      "balance_loss_mlp": 1.01634622,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 1.6385584403278506,
      "language_loss": 0.81732279,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.83932316,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.7425825595855713
    },
    {
      "auxiliary_loss_clip": 0.01181095,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.01502526,
      "balance_loss_mlp": 1.02059722,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 1.9969742454265693,
      "language_loss": 0.81992316,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.84201735,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.64505934715271
    },
    {
      "auxiliary_loss_clip": 0.01182066,
      "auxiliary_loss_mlp": 0.01123904,
      "balance_loss_clip": 1.01661694,
      "balance_loss_mlp": 0.0,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 2.005719871857909,
      "language_loss": 0.71505052,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73811018,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.7250747680664062
    },
    {
      "auxiliary_loss_clip": 0.01184446,
      "auxiliary_loss_mlp": 0.01029546,
      "balance_loss_clip": 1.01316094,
      "balance_loss_mlp": 1.02085543,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 3.1929227480002527,
      "language_loss": 0.83087301,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85301286,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.7316672801971436
    },
    {
      "auxiliary_loss_clip": 0.01185183,
      "auxiliary_loss_mlp": 0.01032922,
      "balance_loss_clip": 0.97806108,
      "balance_loss_mlp": 1.02410698,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 1.5513704456898192,
      "language_loss": 0.83416927,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.8563503,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 3.5576796531677246
    },
    {
      "auxiliary_loss_clip": 0.01178483,
      "auxiliary_loss_mlp": 0.01030418,
      "balance_loss_clip": 1.01284575,
      "balance_loss_mlp": 1.02235365,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 2.1964254473253764,
      "language_loss": 0.79674876,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.81883776,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.731013774871826
    },
    {
      "auxiliary_loss_clip": 0.01187262,
      "auxiliary_loss_mlp": 0.01033206,
      "balance_loss_clip": 1.05587041,
      "balance_loss_mlp": 1.0244863,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 1.8289714963148282,
      "language_loss": 0.8022666,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82447135,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 3.477999210357666
    },
    {
      "auxiliary_loss_clip": 0.0117062,
      "auxiliary_loss_mlp": 0.0103104,
      "balance_loss_clip": 1.01270151,
      "balance_loss_mlp": 1.02310705,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 1.7204685990257371,
      "language_loss": 0.72053087,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.74254751,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.7295708656311035
    },
    {
      "auxiliary_loss_clip": 0.01179704,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.01626229,
      "balance_loss_mlp": 1.02042639,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 2.1699027249472898,
      "language_loss": 0.78114593,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80322701,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.6873841285705566
    },
    {
      "auxiliary_loss_clip": 0.01176276,
      "auxiliary_loss_mlp": 0.0103029,
      "balance_loss_clip": 0.93396628,
      "balance_loss_mlp": 1.02210033,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 1.830403477819858,
      "language_loss": 0.87218171,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.89424735,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.8111095428466797
    },
    {
      "auxiliary_loss_clip": 0.01185333,
      "auxiliary_loss_mlp": 0.0103067,
      "balance_loss_clip": 1.01517105,
      "balance_loss_mlp": 1.02259934,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 1.786084467119313,
      "language_loss": 0.74274522,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.76490521,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.6749398708343506
    },
    {
      "auxiliary_loss_clip": 0.01182348,
      "auxiliary_loss_mlp": 0.01028553,
      "balance_loss_clip": 1.01570618,
      "balance_loss_mlp": 1.02013135,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 2.2422164691391617,
      "language_loss": 0.82500923,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84711826,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.6122946739196777
    },
    {
      "auxiliary_loss_clip": 0.01179625,
      "auxiliary_loss_mlp": 0.01024473,
      "balance_loss_clip": 0.93930167,
      "balance_loss_mlp": 1.01623547,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 1.597851110680576,
      "language_loss": 0.68066919,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70271021,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.752634286880493
    },
    {
      "auxiliary_loss_clip": 0.01183637,
      "auxiliary_loss_mlp": 0.01023891,
      "balance_loss_clip": 1.0521946,
      "balance_loss_mlp": 1.01600814,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 1.8455698670441048,
      "language_loss": 0.74904609,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77112138,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.6603589057922363
    },
    {
      "auxiliary_loss_clip": 0.01181426,
      "auxiliary_loss_mlp": 0.01124456,
      "balance_loss_clip": 0.97787392,
      "balance_loss_mlp": 0.0,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 1.7255970585397673,
      "language_loss": 0.71198219,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.73504108,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 2.7176735401153564
    },
    {
      "auxiliary_loss_clip": 0.01174907,
      "auxiliary_loss_mlp": 0.01022348,
      "balance_loss_clip": 0.93141836,
      "balance_loss_mlp": 1.01426005,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 2.9753636691499414,
      "language_loss": 0.70342004,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.72539258,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 2.702040433883667
    },
    {
      "auxiliary_loss_clip": 0.01167772,
      "auxiliary_loss_mlp": 0.01031356,
      "balance_loss_clip": 0.89621824,
      "balance_loss_mlp": 1.02321398,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 1.6090865191546002,
      "language_loss": 0.70097685,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.72296804,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.9132587909698486
    },
    {
      "auxiliary_loss_clip": 0.01172854,
      "auxiliary_loss_mlp": 0.01033511,
      "balance_loss_clip": 0.97712159,
      "balance_loss_mlp": 1.02562571,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 1.8018461350623733,
      "language_loss": 0.69399405,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.71605772,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 2.857778787612915
    },
    {
      "auxiliary_loss_clip": 0.01178568,
      "auxiliary_loss_mlp": 0.01033128,
      "balance_loss_clip": 0.97540635,
      "balance_loss_mlp": 1.02491474,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.6844729684531932,
      "language_loss": 0.85209835,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.8742153,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 2.7183432579040527
    },
    {
      "auxiliary_loss_clip": 0.01190525,
      "auxiliary_loss_mlp": 0.01033637,
      "balance_loss_clip": 1.01793647,
      "balance_loss_mlp": 1.02560198,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 2.196063519058592,
      "language_loss": 0.74680692,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76904845,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 2.6249353885650635
    },
    {
      "auxiliary_loss_clip": 0.01187008,
      "auxiliary_loss_mlp": 0.0103419,
      "balance_loss_clip": 1.05441213,
      "balance_loss_mlp": 1.02577376,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 4.3623021598367195,
      "language_loss": 0.68965626,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71186829,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 2.5672624111175537
    },
    {
      "auxiliary_loss_clip": 0.0118695,
      "auxiliary_loss_mlp": 0.01030028,
      "balance_loss_clip": 1.01610804,
      "balance_loss_mlp": 1.02186227,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.1095494289547103,
      "language_loss": 0.68593454,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.70810437,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 2.6907007694244385
    },
    {
      "auxiliary_loss_clip": 0.01186829,
      "auxiliary_loss_mlp": 0.0102782,
      "balance_loss_clip": 1.05495405,
      "balance_loss_mlp": 1.01930833,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 2.542910176283857,
      "language_loss": 0.73616183,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75830841,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 2.6223223209381104
    },
    {
      "auxiliary_loss_clip": 0.01185081,
      "auxiliary_loss_mlp": 0.01036824,
      "balance_loss_clip": 1.05439019,
      "balance_loss_mlp": 1.02784133,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 2.8702922796605024,
      "language_loss": 0.66638899,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.68860805,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.6961464881896973
    },
    {
      "auxiliary_loss_clip": 0.01176663,
      "auxiliary_loss_mlp": 0.01027986,
      "balance_loss_clip": 0.93841553,
      "balance_loss_mlp": 1.01928353,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 1.7281813184100023,
      "language_loss": 0.81993091,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.84197736,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 3.67948842048645
    },
    {
      "auxiliary_loss_clip": 0.01185343,
      "auxiliary_loss_mlp": 0.01124209,
      "balance_loss_clip": 1.01491022,
      "balance_loss_mlp": 0.0,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 2.0256014079332534,
      "language_loss": 0.70432442,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.72741997,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 3.5173418521881104
    },
    {
      "auxiliary_loss_clip": 0.01189077,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.05438042,
      "balance_loss_mlp": 1.0201056,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 1.8114806138524737,
      "language_loss": 0.6346159,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65678698,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.6497881412506104
    },
    {
      "auxiliary_loss_clip": 0.01173319,
      "auxiliary_loss_mlp": 0.01033452,
      "balance_loss_clip": 0.97572273,
      "balance_loss_mlp": 1.02520275,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 1.5943132555184534,
      "language_loss": 0.65272492,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67479265,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.645253896713257
    },
    {
      "auxiliary_loss_clip": 0.01175382,
      "auxiliary_loss_mlp": 0.01033814,
      "balance_loss_clip": 0.89938545,
      "balance_loss_mlp": 1.02585101,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.5682447196612825,
      "language_loss": 0.7472744,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.76936638,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.759247303009033
    },
    {
      "auxiliary_loss_clip": 0.01186453,
      "auxiliary_loss_mlp": 0.01027181,
      "balance_loss_clip": 0.98014164,
      "balance_loss_mlp": 1.01906824,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 1.6830542629018956,
      "language_loss": 0.80357945,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.82571578,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.7439253330230713
    },
    {
      "auxiliary_loss_clip": 0.01180211,
      "auxiliary_loss_mlp": 0.01028246,
      "balance_loss_clip": 0.98011583,
      "balance_loss_mlp": 1.0201335,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.0760030119771744,
      "language_loss": 0.68106341,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.70314795,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 3.5506248474121094
    },
    {
      "auxiliary_loss_clip": 0.01182124,
      "auxiliary_loss_mlp": 0.01124138,
      "balance_loss_clip": 0.93856978,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 1.951497147305843,
      "language_loss": 0.80660689,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.82966954,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.7286558151245117
    },
    {
      "auxiliary_loss_clip": 0.01178396,
      "auxiliary_loss_mlp": 0.01029449,
      "balance_loss_clip": 1.01265621,
      "balance_loss_mlp": 1.02126515,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.5408923779362895,
      "language_loss": 0.77671009,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.79878855,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 3.5359303951263428
    },
    {
      "auxiliary_loss_clip": 0.01184829,
      "auxiliary_loss_mlp": 0.01026538,
      "balance_loss_clip": 1.01585925,
      "balance_loss_mlp": 1.01822925,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.1294778126100176,
      "language_loss": 0.79579729,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.81791091,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.66460919380188
    },
    {
      "auxiliary_loss_clip": 0.01179925,
      "auxiliary_loss_mlp": 0.01026933,
      "balance_loss_clip": 1.05235553,
      "balance_loss_mlp": 1.01904106,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 2.0842829868041184,
      "language_loss": 0.86367381,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.88574237,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.6353225708007812
    },
    {
      "auxiliary_loss_clip": 0.01179354,
      "auxiliary_loss_mlp": 0.01032258,
      "balance_loss_clip": 0.98216528,
      "balance_loss_mlp": 1.02427673,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 1.9651772574657023,
      "language_loss": 0.75103402,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77315015,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.7355360984802246
    },
    {
      "auxiliary_loss_clip": 0.01183618,
      "auxiliary_loss_mlp": 0.01026593,
      "balance_loss_clip": 1.05333591,
      "balance_loss_mlp": 1.01827252,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 1.8922469269289623,
      "language_loss": 0.71600753,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73810971,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.7962467670440674
    },
    {
      "auxiliary_loss_clip": 0.01171709,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 0.93803847,
      "balance_loss_mlp": 1.0182755,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 2.0029293233666645,
      "language_loss": 0.56310928,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.58508456,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.8156814575195312
    },
    {
      "auxiliary_loss_clip": 0.01185668,
      "auxiliary_loss_mlp": 0.01026084,
      "balance_loss_clip": 0.93871009,
      "balance_loss_mlp": 1.01804888,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 1.6493037201061789,
      "language_loss": 0.69072419,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71284175,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.7423095703125
    },
    {
      "auxiliary_loss_clip": 0.01187074,
      "auxiliary_loss_mlp": 0.01028374,
      "balance_loss_clip": 1.05606222,
      "balance_loss_mlp": 1.02065825,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 1.8553320949488334,
      "language_loss": 0.73208141,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.75423586,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.600799322128296
    },
    {
      "auxiliary_loss_clip": 0.01182628,
      "auxiliary_loss_mlp": 0.01026349,
      "balance_loss_clip": 0.97658014,
      "balance_loss_mlp": 1.01774812,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 2.0052431814584204,
      "language_loss": 0.77775145,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.79984128,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 2.7686009407043457
    },
    {
      "auxiliary_loss_clip": 0.01187007,
      "auxiliary_loss_mlp": 0.01027327,
      "balance_loss_clip": 0.94031775,
      "balance_loss_mlp": 1.01960194,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.6790099460518757,
      "language_loss": 0.73325312,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.75539649,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 2.7566823959350586
    },
    {
      "auxiliary_loss_clip": 0.01181036,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 0.93727344,
      "balance_loss_mlp": 1.02043271,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 1.7243639091410434,
      "language_loss": 0.70945883,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.73154843,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.719383955001831
    },
    {
      "auxiliary_loss_clip": 0.01181918,
      "auxiliary_loss_mlp": 0.01032003,
      "balance_loss_clip": 1.01486206,
      "balance_loss_mlp": 1.02300286,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 1.9548565196943377,
      "language_loss": 0.82646012,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.84859931,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 2.611849784851074
    },
    {
      "auxiliary_loss_clip": 0.01171589,
      "auxiliary_loss_mlp": 0.01031332,
      "balance_loss_clip": 0.97465456,
      "balance_loss_mlp": 1.02307129,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 1.9650518780893562,
      "language_loss": 0.7558074,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.77783662,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 2.9439823627471924
    },
    {
      "auxiliary_loss_clip": 0.01169569,
      "auxiliary_loss_mlp": 0.010305,
      "balance_loss_clip": 0.85887623,
      "balance_loss_mlp": 1.02216148,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 2.739115032846531,
      "language_loss": 0.7897929,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81179357,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.7714545726776123
    },
    {
      "auxiliary_loss_clip": 0.01187711,
      "auxiliary_loss_mlp": 0.01027004,
      "balance_loss_clip": 1.01648498,
      "balance_loss_mlp": 1.01884985,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.647436510825817,
      "language_loss": 0.68770397,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.70985115,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 2.8416521549224854
    },
    {
      "auxiliary_loss_clip": 0.01191763,
      "auxiliary_loss_mlp": 0.01031373,
      "balance_loss_clip": 0.94380689,
      "balance_loss_mlp": 1.02329087,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 1.4683842584578253,
      "language_loss": 0.71696329,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.73919463,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 2.72100567817688
    },
    {
      "auxiliary_loss_clip": 0.01188003,
      "auxiliary_loss_mlp": 0.0102894,
      "balance_loss_clip": 1.01698697,
      "balance_loss_mlp": 1.02035117,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 2.5258588759431664,
      "language_loss": 0.77162814,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79379749,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.656412124633789
    },
    {
      "auxiliary_loss_clip": 0.0118454,
      "auxiliary_loss_mlp": 0.01124044,
      "balance_loss_clip": 1.05306292,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.0194465605870513,
      "language_loss": 0.88844383,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.91152966,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 3.5131497383117676
    },
    {
      "auxiliary_loss_clip": 0.01184051,
      "auxiliary_loss_mlp": 0.01038526,
      "balance_loss_clip": 0.97525936,
      "balance_loss_mlp": 1.03032494,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.5308363610817732,
      "language_loss": 0.8192879,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.84151363,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.707132339477539
    },
    {
      "auxiliary_loss_clip": 0.0118804,
      "auxiliary_loss_mlp": 0.01027339,
      "balance_loss_clip": 1.01581883,
      "balance_loss_mlp": 1.01907802,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 1.8369661257382301,
      "language_loss": 0.72774088,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.74989468,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 3.815237283706665
    },
    {
      "auxiliary_loss_clip": 0.01180986,
      "auxiliary_loss_mlp": 0.01027129,
      "balance_loss_clip": 0.94150841,
      "balance_loss_mlp": 1.01843286,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 1.8770248795738647,
      "language_loss": 0.84990788,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87198901,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.858339786529541
    },
    {
      "auxiliary_loss_clip": 0.01182152,
      "auxiliary_loss_mlp": 0.01029727,
      "balance_loss_clip": 1.01651549,
      "balance_loss_mlp": 1.02138853,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 2.53678503985766,
      "language_loss": 0.8394866,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86160541,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.650202989578247
    },
    {
      "auxiliary_loss_clip": 0.01183819,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.05494881,
      "balance_loss_mlp": 1.02156818,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 3.1518938239844556,
      "language_loss": 0.78237927,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.80451095,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.639774799346924
    },
    {
      "auxiliary_loss_clip": 0.01181046,
      "auxiliary_loss_mlp": 0.01029586,
      "balance_loss_clip": 1.01508665,
      "balance_loss_mlp": 1.02192092,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 1.6563904534479927,
      "language_loss": 0.65728843,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.67939472,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.741058349609375
    },
    {
      "auxiliary_loss_clip": 0.01167483,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 0.8950069,
      "balance_loss_mlp": 1.0199548,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 2.0946380154339046,
      "language_loss": 0.75053227,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.77248621,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 3.645231246948242
    },
    {
      "auxiliary_loss_clip": 0.01184452,
      "auxiliary_loss_mlp": 0.01030277,
      "balance_loss_clip": 1.01390886,
      "balance_loss_mlp": 1.0223496,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 1.9920172206086932,
      "language_loss": 0.72616172,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.74830902,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.7148256301879883
    },
    {
      "auxiliary_loss_clip": 0.01178245,
      "auxiliary_loss_mlp": 0.0102816,
      "balance_loss_clip": 0.94020748,
      "balance_loss_mlp": 1.02008986,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 1.8730180806672967,
      "language_loss": 0.729195,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75125909,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.6698362827301025
    },
    {
      "auxiliary_loss_clip": 0.01176354,
      "auxiliary_loss_mlp": 0.0103228,
      "balance_loss_clip": 0.90202999,
      "balance_loss_mlp": 1.02378654,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 1.8741162703428256,
      "language_loss": 0.81016481,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83225119,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 3.674501895904541
    },
    {
      "auxiliary_loss_clip": 0.01165927,
      "auxiliary_loss_mlp": 0.01026515,
      "balance_loss_clip": 0.89600819,
      "balance_loss_mlp": 1.01738322,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 1.8361997484755892,
      "language_loss": 0.83239329,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85431767,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.680731773376465
    },
    {
      "auxiliary_loss_clip": 0.01082765,
      "auxiliary_loss_mlp": 0.0100051,
      "balance_loss_clip": 0.94568151,
      "balance_loss_mlp": 0.99847144,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.757556139633834,
      "language_loss": 0.58219266,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60302538,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.4448745250701904
    },
    {
      "auxiliary_loss_clip": 0.01112141,
      "auxiliary_loss_mlp": 0.01001818,
      "balance_loss_clip": 0.89425069,
      "balance_loss_mlp": 0.99974364,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7942360508901716,
      "language_loss": 0.59893233,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.62007189,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.482140302658081
    },
    {
      "auxiliary_loss_clip": 0.01166655,
      "auxiliary_loss_mlp": 0.01029409,
      "balance_loss_clip": 0.8954528,
      "balance_loss_mlp": 1.02153504,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 1.9357242428676125,
      "language_loss": 0.80460685,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.82656753,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.7333786487579346
    },
    {
      "auxiliary_loss_clip": 0.01180732,
      "auxiliary_loss_mlp": 0.0103301,
      "balance_loss_clip": 0.9746629,
      "balance_loss_mlp": 1.02541971,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 2.2110523263100594,
      "language_loss": 0.75163972,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77377707,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.6947226524353027
    },
    {
      "auxiliary_loss_clip": 0.01180269,
      "auxiliary_loss_mlp": 0.01026578,
      "balance_loss_clip": 0.9750998,
      "balance_loss_mlp": 1.01921117,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 1.954353430300777,
      "language_loss": 0.72096181,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74303031,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.7049570083618164
    },
    {
      "auxiliary_loss_clip": 0.01179877,
      "auxiliary_loss_mlp": 0.01027144,
      "balance_loss_clip": 1.05338931,
      "balance_loss_mlp": 1.01955009,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 1.5136994704187066,
      "language_loss": 0.86507386,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.88714409,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 2.6964190006256104
    },
    {
      "auxiliary_loss_clip": 0.01180178,
      "auxiliary_loss_mlp": 0.01030462,
      "balance_loss_clip": 0.97794962,
      "balance_loss_mlp": 1.02250528,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 2.022613735261913,
      "language_loss": 0.7654987,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.78760505,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 2.8018226623535156
    },
    {
      "auxiliary_loss_clip": 0.01175027,
      "auxiliary_loss_mlp": 0.01028066,
      "balance_loss_clip": 0.93763667,
      "balance_loss_mlp": 1.02019858,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 2.20575724883469,
      "language_loss": 0.83869159,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.86072254,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.739224672317505
    },
    {
      "auxiliary_loss_clip": 0.01177744,
      "auxiliary_loss_mlp": 0.01025,
      "balance_loss_clip": 1.01466286,
      "balance_loss_mlp": 1.01724529,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 1.684736097743485,
      "language_loss": 0.82616723,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.84819466,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 2.6818301677703857
    },
    {
      "auxiliary_loss_clip": 0.01186793,
      "auxiliary_loss_mlp": 0.01123951,
      "balance_loss_clip": 1.05566192,
      "balance_loss_mlp": 0.0,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 2.5182625719853053,
      "language_loss": 0.74947381,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.77258134,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 2.6191656589508057
    },
    {
      "auxiliary_loss_clip": 0.01190076,
      "auxiliary_loss_mlp": 0.01029009,
      "balance_loss_clip": 0.97947669,
      "balance_loss_mlp": 1.02065253,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 1.8322138081494195,
      "language_loss": 0.83946031,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86165118,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 2.718862771987915
    },
    {
      "auxiliary_loss_clip": 0.0117111,
      "auxiliary_loss_mlp": 0.0102998,
      "balance_loss_clip": 0.97470117,
      "balance_loss_mlp": 1.02242839,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 1.797255797172715,
      "language_loss": 0.76901805,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.79102898,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 2.6821842193603516
    },
    {
      "auxiliary_loss_clip": 0.01182542,
      "auxiliary_loss_mlp": 0.01028975,
      "balance_loss_clip": 1.05219114,
      "balance_loss_mlp": 1.02095771,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 2.727246998783956,
      "language_loss": 0.77932179,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.80143696,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 2.717923879623413
    },
    {
      "auxiliary_loss_clip": 0.0118116,
      "auxiliary_loss_mlp": 0.01028517,
      "balance_loss_clip": 1.01225626,
      "balance_loss_mlp": 1.02048826,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 1.9076046784734355,
      "language_loss": 0.7885679,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.81066465,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.696209192276001
    },
    {
      "auxiliary_loss_clip": 0.01178048,
      "auxiliary_loss_mlp": 0.01036474,
      "balance_loss_clip": 0.97500682,
      "balance_loss_mlp": 1.02826691,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 1.7675181886146214,
      "language_loss": 0.73284364,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.75498879,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 2.7591705322265625
    },
    {
      "auxiliary_loss_clip": 0.01185218,
      "auxiliary_loss_mlp": 0.0102692,
      "balance_loss_clip": 1.01498759,
      "balance_loss_mlp": 1.01874804,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.4179567765607057,
      "language_loss": 0.87104797,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.8931694,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 3.8302388191223145
    },
    {
      "auxiliary_loss_clip": 0.01176192,
      "auxiliary_loss_mlp": 0.01031574,
      "balance_loss_clip": 1.01444161,
      "balance_loss_mlp": 1.02324688,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 1.6964378421711737,
      "language_loss": 0.81725961,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.83933723,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 3.6259963512420654
    },
    {
      "auxiliary_loss_clip": 0.01181276,
      "auxiliary_loss_mlp": 0.01021519,
      "balance_loss_clip": 1.01629567,
      "balance_loss_mlp": 1.01436329,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 1.7071266815464243,
      "language_loss": 0.76083744,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.78286535,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.7170488834381104
    },
    {
      "auxiliary_loss_clip": 0.01167757,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 0.97111928,
      "balance_loss_mlp": 1.01981246,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 2.092930645881514,
      "language_loss": 0.78775007,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.80970305,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.71307373046875
    },
    {
      "auxiliary_loss_clip": 0.01170068,
      "auxiliary_loss_mlp": 0.01029498,
      "balance_loss_clip": 0.97289622,
      "balance_loss_mlp": 1.02210057,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 1.5266628215455897,
      "language_loss": 0.83469534,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.856691,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.7430338859558105
    },
    {
      "auxiliary_loss_clip": 0.01184002,
      "auxiliary_loss_mlp": 0.01027673,
      "balance_loss_clip": 1.05453515,
      "balance_loss_mlp": 1.01964974,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 3.4101130198121528,
      "language_loss": 0.67654562,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.6986624,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.57377552986145
    },
    {
      "auxiliary_loss_clip": 0.01181438,
      "auxiliary_loss_mlp": 0.01028077,
      "balance_loss_clip": 1.05253458,
      "balance_loss_mlp": 1.02010775,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 5.2528813312950575,
      "language_loss": 0.72397166,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.74606681,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 3.518630266189575
    },
    {
      "auxiliary_loss_clip": 0.01183592,
      "auxiliary_loss_mlp": 0.01027382,
      "balance_loss_clip": 0.9358784,
      "balance_loss_mlp": 1.01931739,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 1.5235642596089882,
      "language_loss": 0.68074322,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70285296,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.9102931022644043
    },
    {
      "auxiliary_loss_clip": 0.01172048,
      "auxiliary_loss_mlp": 0.01025792,
      "balance_loss_clip": 0.89670515,
      "balance_loss_mlp": 1.01784337,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 2.7033393566748707,
      "language_loss": 0.76895791,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.79093635,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.8125078678131104
    },
    {
      "auxiliary_loss_clip": 0.01179091,
      "auxiliary_loss_mlp": 0.01024007,
      "balance_loss_clip": 0.97507083,
      "balance_loss_mlp": 1.01612687,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 3.05690023807575,
      "language_loss": 0.85200089,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.8740319,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 3.5374696254730225
    },
    {
      "auxiliary_loss_clip": 0.01172419,
      "auxiliary_loss_mlp": 0.01034101,
      "balance_loss_clip": 0.93813765,
      "balance_loss_mlp": 1.02575064,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 1.9855345328704697,
      "language_loss": 0.80818081,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83024597,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.707895278930664
    },
    {
      "auxiliary_loss_clip": 0.0118442,
      "auxiliary_loss_mlp": 0.01029659,
      "balance_loss_clip": 0.89904362,
      "balance_loss_mlp": 1.02141011,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 1.634839588448174,
      "language_loss": 0.70763707,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.72977787,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.763254404067993
    },
    {
      "auxiliary_loss_clip": 0.01185747,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 0.90239143,
      "balance_loss_mlp": 1.0194664,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.7280738588484823,
      "language_loss": 0.82647538,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.84860492,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.7682788372039795
    },
    {
      "auxiliary_loss_clip": 0.01164189,
      "auxiliary_loss_mlp": 0.0103013,
      "balance_loss_clip": 0.93572003,
      "balance_loss_mlp": 1.02255726,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 2.645491829639233,
      "language_loss": 0.82470715,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84665036,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.7120211124420166
    },
    {
      "auxiliary_loss_clip": 0.01179009,
      "auxiliary_loss_mlp": 0.01029518,
      "balance_loss_clip": 0.81815588,
      "balance_loss_mlp": 1.02199602,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 2.3229659988369473,
      "language_loss": 0.83398157,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.85606682,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 3.026613473892212
    },
    {
      "auxiliary_loss_clip": 0.01166353,
      "auxiliary_loss_mlp": 0.01027716,
      "balance_loss_clip": 0.89475721,
      "balance_loss_mlp": 1.01991928,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 1.6123315591103666,
      "language_loss": 0.77631915,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.79825985,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 3.0080325603485107
    },
    {
      "auxiliary_loss_clip": 0.01176581,
      "auxiliary_loss_mlp": 0.01028581,
      "balance_loss_clip": 1.01088023,
      "balance_loss_mlp": 1.02000427,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 2.428444235513101,
      "language_loss": 0.78428912,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80634069,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 2.643538475036621
    },
    {
      "auxiliary_loss_clip": 0.01080021,
      "auxiliary_loss_mlp": 0.01003499,
      "balance_loss_clip": 1.01912808,
      "balance_loss_mlp": 1.00155592,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.771385286715895,
      "language_loss": 0.62519717,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.64603233,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.3185129165649414
    },
    {
      "auxiliary_loss_clip": 0.01183299,
      "auxiliary_loss_mlp": 0.01030194,
      "balance_loss_clip": 0.93861496,
      "balance_loss_mlp": 1.02199888,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.625981328626022,
      "language_loss": 0.84483886,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.86697376,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 2.724113702774048
    },
    {
      "auxiliary_loss_clip": 0.01173034,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 0.97436631,
      "balance_loss_mlp": 1.02087927,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 1.8887400787345745,
      "language_loss": 0.81854576,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.84056813,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 2.7610223293304443
    },
    {
      "auxiliary_loss_clip": 0.01176414,
      "auxiliary_loss_mlp": 0.01028999,
      "balance_loss_clip": 0.93741524,
      "balance_loss_mlp": 1.02054095,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 1.963045125444464,
      "language_loss": 0.74219,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.7642442,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 2.761396884918213
    },
    {
      "auxiliary_loss_clip": 0.01084652,
      "auxiliary_loss_mlp": 0.01002734,
      "balance_loss_clip": 0.981924,
      "balance_loss_mlp": 1.00071895,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.814259033285699,
      "language_loss": 0.62914717,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.65002102,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 3.23954439163208
    },
    {
      "auxiliary_loss_clip": 0.01179392,
      "auxiliary_loss_mlp": 0.01033082,
      "balance_loss_clip": 0.97793651,
      "balance_loss_mlp": 1.0252471,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.1129719867907237,
      "language_loss": 0.71841812,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74054289,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 2.7471721172332764
    },
    {
      "auxiliary_loss_clip": 0.01180734,
      "auxiliary_loss_mlp": 0.01025514,
      "balance_loss_clip": 1.05368423,
      "balance_loss_mlp": 1.01863837,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 1.9348189428637648,
      "language_loss": 0.80670774,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.82877022,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 2.6620969772338867
    },
    {
      "auxiliary_loss_clip": 0.01180401,
      "auxiliary_loss_mlp": 0.01028986,
      "balance_loss_clip": 1.01281738,
      "balance_loss_mlp": 1.02130318,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 1.79765597249183,
      "language_loss": 0.7324068,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.75450069,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 2.681292772293091
    },
    {
      "auxiliary_loss_clip": 0.01174206,
      "auxiliary_loss_mlp": 0.01033084,
      "balance_loss_clip": 1.01055384,
      "balance_loss_mlp": 1.02560937,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 1.9924639994011841,
      "language_loss": 0.76079977,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78287268,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 2.6154708862304688
    },
    {
      "auxiliary_loss_clip": 0.01174908,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 1.01126766,
      "balance_loss_mlp": 1.01776671,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 1.8370350408690994,
      "language_loss": 0.89448392,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.9164831,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 3.601923942565918
    },
    {
      "auxiliary_loss_clip": 0.0118258,
      "auxiliary_loss_mlp": 0.01024672,
      "balance_loss_clip": 1.01598489,
      "balance_loss_mlp": 1.01743555,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 1.6012605759970084,
      "language_loss": 0.80153668,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.82360917,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.6866469383239746
    },
    {
      "auxiliary_loss_clip": 0.0118103,
      "auxiliary_loss_mlp": 0.01030822,
      "balance_loss_clip": 0.9738512,
      "balance_loss_mlp": 1.02328777,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 2.623813919362318,
      "language_loss": 0.7425971,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76471567,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 3.6192190647125244
    },
    {
      "auxiliary_loss_clip": 0.01178476,
      "auxiliary_loss_mlp": 0.0102758,
      "balance_loss_clip": 1.05216217,
      "balance_loss_mlp": 1.01984358,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 1.6721595430171536,
      "language_loss": 0.76478821,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78684878,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.755356550216675
    },
    {
      "auxiliary_loss_clip": 0.01177901,
      "auxiliary_loss_mlp": 0.01123434,
      "balance_loss_clip": 1.01581383,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 1.977362836268463,
      "language_loss": 0.8189851,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.84199846,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.732628107070923
    },
    {
      "auxiliary_loss_clip": 0.01173409,
      "auxiliary_loss_mlp": 0.01033474,
      "balance_loss_clip": 1.00949621,
      "balance_loss_mlp": 1.02536178,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.104302853838631,
      "language_loss": 0.6585114,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68058026,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.6326348781585693
    },
    {
      "auxiliary_loss_clip": 0.01179298,
      "auxiliary_loss_mlp": 0.0102785,
      "balance_loss_clip": 1.01121867,
      "balance_loss_mlp": 1.02005649,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 2.0416618418645833,
      "language_loss": 0.7940591,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81613058,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 3.6234633922576904
    },
    {
      "auxiliary_loss_clip": 0.01168149,
      "auxiliary_loss_mlp": 0.01029293,
      "balance_loss_clip": 0.97250295,
      "balance_loss_mlp": 1.02144909,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.8114792469600445,
      "language_loss": 0.85836446,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.88033885,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.6263558864593506
    },
    {
      "auxiliary_loss_clip": 0.01175759,
      "auxiliary_loss_mlp": 0.01027465,
      "balance_loss_clip": 0.89407915,
      "balance_loss_mlp": 1.01943326,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 3.349273103305113,
      "language_loss": 0.67924607,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.70127827,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.73378849029541
    },
    {
      "auxiliary_loss_clip": 0.0117448,
      "auxiliary_loss_mlp": 0.01031975,
      "balance_loss_clip": 0.97673541,
      "balance_loss_mlp": 1.02510285,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.7509166245772794,
      "language_loss": 0.82141221,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84347671,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 3.525686502456665
    },
    {
      "auxiliary_loss_clip": 0.01176091,
      "auxiliary_loss_mlp": 0.01122935,
      "balance_loss_clip": 1.01410842,
      "balance_loss_mlp": 0.0,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 1.8978204377093464,
      "language_loss": 0.67086655,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.69385684,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.636038064956665
    },
    {
      "auxiliary_loss_clip": 0.01169864,
      "auxiliary_loss_mlp": 0.01029451,
      "balance_loss_clip": 0.93456835,
      "balance_loss_mlp": 1.02153516,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 1.9965396493051664,
      "language_loss": 0.7936511,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81564426,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.6387083530426025
    },
    {
      "auxiliary_loss_clip": 0.01173334,
      "auxiliary_loss_mlp": 0.0102941,
      "balance_loss_clip": 0.97460777,
      "balance_loss_mlp": 1.02092862,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 1.8386522465780135,
      "language_loss": 0.72925639,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75128382,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.6676864624023438
    },
    {
      "auxiliary_loss_clip": 0.01174323,
      "auxiliary_loss_mlp": 0.0102316,
      "balance_loss_clip": 1.00951171,
      "balance_loss_mlp": 1.0158819,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 1.9309957082488196,
      "language_loss": 0.61393774,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63591254,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.778479814529419
    },
    {
      "auxiliary_loss_clip": 0.01173362,
      "auxiliary_loss_mlp": 0.01030919,
      "balance_loss_clip": 1.01381052,
      "balance_loss_mlp": 1.02340257,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 1.7989216225019589,
      "language_loss": 0.78841925,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.810462,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.663454532623291
    },
    {
      "auxiliary_loss_clip": 0.01173835,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.01421821,
      "balance_loss_mlp": 1.02181995,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.6447452567881582,
      "language_loss": 0.81297576,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.83500487,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.699350595474243
    },
    {
      "auxiliary_loss_clip": 0.01179904,
      "auxiliary_loss_mlp": 0.01028717,
      "balance_loss_clip": 1.0154109,
      "balance_loss_mlp": 1.02112579,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 1.5536617547583322,
      "language_loss": 0.72573066,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.74781686,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 2.6544077396392822
    },
    {
      "auxiliary_loss_clip": 0.0118337,
      "auxiliary_loss_mlp": 0.01123649,
      "balance_loss_clip": 1.01613283,
      "balance_loss_mlp": 0.0,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 1.9336113757379765,
      "language_loss": 0.91217017,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.93524033,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.6972672939300537
    },
    {
      "auxiliary_loss_clip": 0.01170425,
      "auxiliary_loss_mlp": 0.01030452,
      "balance_loss_clip": 1.01189065,
      "balance_loss_mlp": 1.02292991,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 1.85566492689334,
      "language_loss": 0.76011354,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78212237,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 2.6926541328430176
    },
    {
      "auxiliary_loss_clip": 0.01177091,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.01385999,
      "balance_loss_mlp": 1.02162671,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.4483816368283435,
      "language_loss": 0.78491867,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.80698383,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 2.679877519607544
    },
    {
      "auxiliary_loss_clip": 0.0117866,
      "auxiliary_loss_mlp": 0.01025085,
      "balance_loss_clip": 0.9735986,
      "balance_loss_mlp": 1.01814151,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 1.8471937957093116,
      "language_loss": 0.63736141,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.65939885,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 2.7589542865753174
    },
    {
      "auxiliary_loss_clip": 0.01180021,
      "auxiliary_loss_mlp": 0.01033823,
      "balance_loss_clip": 0.93713498,
      "balance_loss_mlp": 1.02541327,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 1.8846299051227082,
      "language_loss": 0.76873159,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.79087007,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 2.7265403270721436
    },
    {
      "auxiliary_loss_clip": 0.01173019,
      "auxiliary_loss_mlp": 0.01123239,
      "balance_loss_clip": 0.93507624,
      "balance_loss_mlp": 0.0,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 1.534455471003871,
      "language_loss": 0.75304163,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.7760042,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.739321708679199
    },
    {
      "auxiliary_loss_clip": 0.01183399,
      "auxiliary_loss_mlp": 0.01032331,
      "balance_loss_clip": 1.01522005,
      "balance_loss_mlp": 1.024737,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.4560709931654807,
      "language_loss": 0.74774039,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.76989764,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 2.6811909675598145
    },
    {
      "auxiliary_loss_clip": 0.01184181,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 1.05314219,
      "balance_loss_mlp": 1.02024221,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 1.643555635614546,
      "language_loss": 0.72730941,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.74943405,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 2.5957493782043457
    },
    {
      "auxiliary_loss_clip": 0.01178312,
      "auxiliary_loss_mlp": 0.01027561,
      "balance_loss_clip": 1.01247334,
      "balance_loss_mlp": 1.01917481,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.44264830705297,
      "language_loss": 0.67709732,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.69915599,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 2.858718156814575
    },
    {
      "auxiliary_loss_clip": 0.01179842,
      "auxiliary_loss_mlp": 0.01023653,
      "balance_loss_clip": 1.05314279,
      "balance_loss_mlp": 1.01626837,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.4040279755884395,
      "language_loss": 0.7984345,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.8204695,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 3.533043146133423
    },
    {
      "auxiliary_loss_clip": 0.01181608,
      "auxiliary_loss_mlp": 0.01036479,
      "balance_loss_clip": 1.05275369,
      "balance_loss_mlp": 1.02847743,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 1.8901311153155127,
      "language_loss": 0.74194312,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.76412392,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 3.6352312564849854
    },
    {
      "auxiliary_loss_clip": 0.01176706,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 0.97600454,
      "balance_loss_mlp": 1.01873064,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.612416944800895,
      "language_loss": 0.63433367,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65636271,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 2.7769525051116943
    },
    {
      "auxiliary_loss_clip": 0.0117975,
      "auxiliary_loss_mlp": 0.01024046,
      "balance_loss_clip": 1.05262589,
      "balance_loss_mlp": 1.01617265,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 2.0637409973088148,
      "language_loss": 0.83121943,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.85325742,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.5894248485565186
    },
    {
      "auxiliary_loss_clip": 0.0117857,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.01517141,
      "balance_loss_mlp": 1.0208776,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 1.5269116758956032,
      "language_loss": 0.78631234,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.80838156,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.6445746421813965
    },
    {
      "auxiliary_loss_clip": 0.01183132,
      "auxiliary_loss_mlp": 0.01025348,
      "balance_loss_clip": 1.01485276,
      "balance_loss_mlp": 1.01784039,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 1.6575589868465423,
      "language_loss": 0.82083011,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84291488,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.720424175262451
    },
    {
      "auxiliary_loss_clip": 0.01175413,
      "auxiliary_loss_mlp": 0.01031005,
      "balance_loss_clip": 0.85972106,
      "balance_loss_mlp": 1.02223134,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 2.035475552656223,
      "language_loss": 0.74224776,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76431191,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.82926869392395
    },
    {
      "auxiliary_loss_clip": 0.01179536,
      "auxiliary_loss_mlp": 0.01026358,
      "balance_loss_clip": 1.05099726,
      "balance_loss_mlp": 1.01868689,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 1.940931909905118,
      "language_loss": 0.71555674,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.7376157,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 3.525672674179077
    },
    {
      "auxiliary_loss_clip": 0.01174013,
      "auxiliary_loss_mlp": 0.01023784,
      "balance_loss_clip": 0.97324026,
      "balance_loss_mlp": 1.01591671,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 1.7172367691297787,
      "language_loss": 0.79030645,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81228435,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.6665539741516113
    },
    {
      "auxiliary_loss_clip": 0.01177491,
      "auxiliary_loss_mlp": 0.01025579,
      "balance_loss_clip": 0.97490633,
      "balance_loss_mlp": 1.01836085,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.520723755385773,
      "language_loss": 0.81422281,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83625352,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.7157576084136963
    },
    {
      "auxiliary_loss_clip": 0.01160229,
      "auxiliary_loss_mlp": 0.01024303,
      "balance_loss_clip": 0.89472032,
      "balance_loss_mlp": 1.01642609,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.6505001281074074,
      "language_loss": 0.7886101,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.81045538,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 3.7002596855163574
    },
    {
      "auxiliary_loss_clip": 0.01174467,
      "auxiliary_loss_mlp": 0.01033203,
      "balance_loss_clip": 0.93767905,
      "balance_loss_mlp": 1.02518618,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.6086064732699694,
      "language_loss": 0.75803936,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78011608,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.801769256591797
    },
    {
      "auxiliary_loss_clip": 0.01182144,
      "auxiliary_loss_mlp": 0.0112343,
      "balance_loss_clip": 1.05134797,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 2.0412208981252786,
      "language_loss": 0.80335838,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.82641417,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.6210222244262695
    },
    {
      "auxiliary_loss_clip": 0.0118292,
      "auxiliary_loss_mlp": 0.01027274,
      "balance_loss_clip": 1.0541656,
      "balance_loss_mlp": 1.0193522,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 3.6249815121461313,
      "language_loss": 0.68082917,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.70293111,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.5650651454925537
    },
    {
      "auxiliary_loss_clip": 0.01177962,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 0.93824345,
      "balance_loss_mlp": 1.01967752,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 1.6019845872073553,
      "language_loss": 0.72181153,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.74386632,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.700230360031128
    },
    {
      "auxiliary_loss_clip": 0.01178726,
      "auxiliary_loss_mlp": 0.01031097,
      "balance_loss_clip": 0.9798007,
      "balance_loss_mlp": 1.02342618,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 2.107848890856697,
      "language_loss": 0.84395373,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.86605197,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.6494975090026855
    },
    {
      "auxiliary_loss_clip": 0.0117987,
      "auxiliary_loss_mlp": 0.0102756,
      "balance_loss_clip": 1.01301479,
      "balance_loss_mlp": 1.01954317,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 1.8450470312974536,
      "language_loss": 0.64230347,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.66437769,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 2.6700918674468994
    },
    {
      "auxiliary_loss_clip": 0.01086002,
      "auxiliary_loss_mlp": 0.0100038,
      "balance_loss_clip": 0.90923887,
      "balance_loss_mlp": 0.9983061,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.8089513541824346,
      "language_loss": 0.56540263,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58626646,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.273853063583374
    },
    {
      "auxiliary_loss_clip": 0.01186058,
      "auxiliary_loss_mlp": 0.01035753,
      "balance_loss_clip": 0.97981137,
      "balance_loss_mlp": 1.02698541,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 1.9805480198256198,
      "language_loss": 0.73365009,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.7558682,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.707585334777832
    },
    {
      "auxiliary_loss_clip": 0.01177888,
      "auxiliary_loss_mlp": 0.01027248,
      "balance_loss_clip": 0.97458982,
      "balance_loss_mlp": 1.01960373,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 1.617074456626834,
      "language_loss": 0.78637004,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.80842143,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 2.7523419857025146
    },
    {
      "auxiliary_loss_clip": 0.01186918,
      "auxiliary_loss_mlp": 0.01029636,
      "balance_loss_clip": 0.97672123,
      "balance_loss_mlp": 1.02129722,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 2.824599554424758,
      "language_loss": 0.80031216,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.82247764,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 2.751363754272461
    },
    {
      "auxiliary_loss_clip": 0.01178052,
      "auxiliary_loss_mlp": 0.01028657,
      "balance_loss_clip": 1.01390576,
      "balance_loss_mlp": 1.02043128,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 1.7447117437975892,
      "language_loss": 0.80029643,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.82236356,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 2.67191219329834
    },
    {
      "auxiliary_loss_clip": 0.01187012,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.01628947,
      "balance_loss_mlp": 1.01931262,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 2.6015197654744537,
      "language_loss": 0.78868127,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.81083089,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 2.739126682281494
    },
    {
      "auxiliary_loss_clip": 0.01182461,
      "auxiliary_loss_mlp": 0.0103312,
      "balance_loss_clip": 1.05574846,
      "balance_loss_mlp": 1.0249902,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.7058062382925838,
      "language_loss": 0.80353338,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.8256892,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 2.72830867767334
    },
    {
      "auxiliary_loss_clip": 0.01177779,
      "auxiliary_loss_mlp": 0.0102183,
      "balance_loss_clip": 1.01285326,
      "balance_loss_mlp": 1.01415324,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 1.3405803556231821,
      "language_loss": 0.80055326,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82254934,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 2.75921630859375
    },
    {
      "auxiliary_loss_clip": 0.01186266,
      "auxiliary_loss_mlp": 0.01035082,
      "balance_loss_clip": 1.05601215,
      "balance_loss_mlp": 1.02724051,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 2.2181030830447375,
      "language_loss": 0.71184921,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.73406267,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 2.664879083633423
    },
    {
      "auxiliary_loss_clip": 0.01179148,
      "auxiliary_loss_mlp": 0.01025459,
      "balance_loss_clip": 1.01410675,
      "balance_loss_mlp": 1.01695359,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 2.104669264617669,
      "language_loss": 0.73298478,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75503087,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 3.5666141510009766
    },
    {
      "auxiliary_loss_clip": 0.01182795,
      "auxiliary_loss_mlp": 0.01025736,
      "balance_loss_clip": 0.97615159,
      "balance_loss_mlp": 1.01779652,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 1.9340150829431766,
      "language_loss": 0.82367969,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84576499,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 3.6382944583892822
    },
    {
      "auxiliary_loss_clip": 0.01180713,
      "auxiliary_loss_mlp": 0.01025551,
      "balance_loss_clip": 0.979114,
      "balance_loss_mlp": 1.01803446,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 1.7157077288028184,
      "language_loss": 0.63841796,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66048062,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.7329397201538086
    },
    {
      "auxiliary_loss_clip": 0.01184238,
      "auxiliary_loss_mlp": 0.01123051,
      "balance_loss_clip": 1.05419731,
      "balance_loss_mlp": 0.0,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 1.9566870451750777,
      "language_loss": 0.79908001,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.82215297,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.692387580871582
    },
    {
      "auxiliary_loss_clip": 0.01181712,
      "auxiliary_loss_mlp": 0.01027651,
      "balance_loss_clip": 1.0540005,
      "balance_loss_mlp": 1.0200007,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 2.772793531920568,
      "language_loss": 0.82634419,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.84843779,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.6504337787628174
    },
    {
      "auxiliary_loss_clip": 0.01189172,
      "auxiliary_loss_mlp": 0.01027363,
      "balance_loss_clip": 0.93817371,
      "balance_loss_mlp": 1.01986468,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 1.8057579586768553,
      "language_loss": 0.86091578,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.8830812,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.7247140407562256
    },
    {
      "auxiliary_loss_clip": 0.01175279,
      "auxiliary_loss_mlp": 0.01123291,
      "balance_loss_clip": 0.8962661,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 1.7103313949492143,
      "language_loss": 0.65183961,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.67482531,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.8262157440185547
    },
    {
      "auxiliary_loss_clip": 0.01183663,
      "auxiliary_loss_mlp": 0.01033628,
      "balance_loss_clip": 1.01352334,
      "balance_loss_mlp": 1.02589142,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 1.6336300995570627,
      "language_loss": 0.79917753,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82135046,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 4.222455739974976
    },
    {
      "auxiliary_loss_clip": 0.01183808,
      "auxiliary_loss_mlp": 0.01028442,
      "balance_loss_clip": 1.0547719,
      "balance_loss_mlp": 1.02032351,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 1.5639861707146598,
      "language_loss": 0.76008672,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78220928,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.7918498516082764
    },
    {
      "auxiliary_loss_clip": 0.01175423,
      "auxiliary_loss_mlp": 0.01027504,
      "balance_loss_clip": 1.01088238,
      "balance_loss_mlp": 1.02054787,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 1.8634826245142948,
      "language_loss": 0.68056327,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.70259249,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.7514243125915527
    },
    {
      "auxiliary_loss_clip": 0.01174054,
      "auxiliary_loss_mlp": 0.01027076,
      "balance_loss_clip": 1.0134536,
      "balance_loss_mlp": 1.01938701,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 1.6621334024143253,
      "language_loss": 0.79213738,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81414866,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 3.5696890354156494
    },
    {
      "auxiliary_loss_clip": 0.01178379,
      "auxiliary_loss_mlp": 0.01023343,
      "balance_loss_clip": 0.94169497,
      "balance_loss_mlp": 1.01561821,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 1.5162524952443193,
      "language_loss": 0.70729095,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72930825,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.8019964694976807
    },
    {
      "auxiliary_loss_clip": 0.01179738,
      "auxiliary_loss_mlp": 0.01030361,
      "balance_loss_clip": 0.90005791,
      "balance_loss_mlp": 1.02225471,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 2.9576822817120982,
      "language_loss": 0.74090487,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.76300585,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.79168438911438
    },
    {
      "auxiliary_loss_clip": 0.01182814,
      "auxiliary_loss_mlp": 0.01123158,
      "balance_loss_clip": 0.93746746,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 1.8759965885794865,
      "language_loss": 0.81357157,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.8366313,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.7377727031707764
    },
    {
      "auxiliary_loss_clip": 0.01181889,
      "auxiliary_loss_mlp": 0.010268,
      "balance_loss_clip": 0.93760145,
      "balance_loss_mlp": 1.01867568,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 1.9949998302790641,
      "language_loss": 0.83807933,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86016631,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.7102699279785156
    },
    {
      "auxiliary_loss_clip": 0.01170522,
      "auxiliary_loss_mlp": 0.01025505,
      "balance_loss_clip": 0.97149736,
      "balance_loss_mlp": 1.01745212,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.8185153801245773,
      "language_loss": 0.71959275,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74155301,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.667665719985962
    },
    {
      "auxiliary_loss_clip": 0.01099413,
      "auxiliary_loss_mlp": 0.01003249,
      "balance_loss_clip": 0.85013974,
      "balance_loss_mlp": 1.00118625,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7495430508026236,
      "language_loss": 0.6374656,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65849221,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 3.416773557662964
    },
    {
      "auxiliary_loss_clip": 0.01184482,
      "auxiliary_loss_mlp": 0.01032202,
      "balance_loss_clip": 1.05430508,
      "balance_loss_mlp": 1.02435255,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 1.8718279799183892,
      "language_loss": 0.76082563,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78299248,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.642162561416626
    },
    {
      "auxiliary_loss_clip": 0.01178949,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 0.97393447,
      "balance_loss_mlp": 1.02185464,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 2.337249096009304,
      "language_loss": 0.81730533,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.83939075,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.6722962856292725
    },
    {
      "auxiliary_loss_clip": 0.01176936,
      "auxiliary_loss_mlp": 0.01033699,
      "balance_loss_clip": 0.97658896,
      "balance_loss_mlp": 1.0262661,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 1.7114521730332501,
      "language_loss": 0.85614586,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.87825227,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 2.658738851547241
    },
    {
      "auxiliary_loss_clip": 0.01179037,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.05221808,
      "balance_loss_mlp": 1.01932847,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 1.9748417614280838,
      "language_loss": 0.78385568,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80591714,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 2.6171867847442627
    },
    {
      "auxiliary_loss_clip": 0.01080856,
      "auxiliary_loss_mlp": 0.01000652,
      "balance_loss_clip": 1.02072859,
      "balance_loss_mlp": 0.99868548,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.7945541631189439,
      "language_loss": 0.62641823,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.64723337,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 3.167912483215332
    },
    {
      "auxiliary_loss_clip": 0.01185556,
      "auxiliary_loss_mlp": 0.01032709,
      "balance_loss_clip": 0.97905451,
      "balance_loss_mlp": 1.02484739,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 1.696415590811519,
      "language_loss": 0.84958357,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87176621,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.7269718647003174
    },
    {
      "auxiliary_loss_clip": 0.01171392,
      "auxiliary_loss_mlp": 0.01025827,
      "balance_loss_clip": 0.93133169,
      "balance_loss_mlp": 1.01754785,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 1.9009701192365227,
      "language_loss": 0.69104612,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.71301824,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.71795916557312
    },
    {
      "auxiliary_loss_clip": 0.01171923,
      "auxiliary_loss_mlp": 0.01028437,
      "balance_loss_clip": 0.97360671,
      "balance_loss_mlp": 1.02130222,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 1.8801183003646522,
      "language_loss": 0.82323337,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.84523696,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 2.6356818675994873
    },
    {
      "auxiliary_loss_clip": 0.01179983,
      "auxiliary_loss_mlp": 0.01026679,
      "balance_loss_clip": 1.01258671,
      "balance_loss_mlp": 1.01902008,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 2.062357111743258,
      "language_loss": 0.67628932,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.69835603,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 3.745187997817993
    },
    {
      "auxiliary_loss_clip": 0.01161428,
      "auxiliary_loss_mlp": 0.01026109,
      "balance_loss_clip": 0.93746346,
      "balance_loss_mlp": 1.01910532,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 1.7930922023183729,
      "language_loss": 0.72551656,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.74739194,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 2.7598681449890137
    },
    {
      "auxiliary_loss_clip": 0.01081199,
      "auxiliary_loss_mlp": 0.01005474,
      "balance_loss_clip": 1.02110338,
      "balance_loss_mlp": 1.00350749,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.8461277284827895,
      "language_loss": 0.55274415,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.5736109,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 4.316458463668823
    },
    {
      "auxiliary_loss_clip": 0.01177689,
      "auxiliary_loss_mlp": 0.0102866,
      "balance_loss_clip": 0.97579414,
      "balance_loss_mlp": 1.02047026,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 2.75859660414755,
      "language_loss": 0.66309249,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68515599,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 2.6829447746276855
    },
    {
      "auxiliary_loss_clip": 0.01179999,
      "auxiliary_loss_mlp": 0.01032858,
      "balance_loss_clip": 1.01536798,
      "balance_loss_mlp": 1.02488542,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.0734740365897126,
      "language_loss": 0.85426891,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87639749,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.688155174255371
    },
    {
      "auxiliary_loss_clip": 0.01175516,
      "auxiliary_loss_mlp": 0.01031227,
      "balance_loss_clip": 0.89951462,
      "balance_loss_mlp": 1.02346075,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 2.090673568173791,
      "language_loss": 0.66926968,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69133711,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.836580991744995
    },
    {
      "auxiliary_loss_clip": 0.01177257,
      "auxiliary_loss_mlp": 0.01027284,
      "balance_loss_clip": 0.97475123,
      "balance_loss_mlp": 1.01939797,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 1.9761821539613176,
      "language_loss": 0.6364429,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.65848827,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.766446828842163
    },
    {
      "auxiliary_loss_clip": 0.01169738,
      "auxiliary_loss_mlp": 0.01034475,
      "balance_loss_clip": 0.97432017,
      "balance_loss_mlp": 1.02740288,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 1.766693225748727,
      "language_loss": 0.71425879,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73630095,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 3.6623754501342773
    },
    {
      "auxiliary_loss_clip": 0.01181733,
      "auxiliary_loss_mlp": 0.01023871,
      "balance_loss_clip": 1.0146687,
      "balance_loss_mlp": 1.01661062,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 1.7534684789792143,
      "language_loss": 0.80641788,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.82847393,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.7025582790374756
    },
    {
      "auxiliary_loss_clip": 0.01178688,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.05443573,
      "balance_loss_mlp": 1.01717162,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 1.6141070808389437,
      "language_loss": 0.72103786,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74307311,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.6742091178894043
    },
    {
      "auxiliary_loss_clip": 0.01183259,
      "auxiliary_loss_mlp": 0.01024596,
      "balance_loss_clip": 1.05374944,
      "balance_loss_mlp": 1.01707423,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 3.7545619142271014,
      "language_loss": 0.79356098,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.8156395,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.646331548690796
    },
    {
      "auxiliary_loss_clip": 0.01082452,
      "auxiliary_loss_mlp": 0.01002705,
      "balance_loss_clip": 0.98199815,
      "balance_loss_mlp": 1.00060654,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.9973391865312368,
      "language_loss": 0.67553055,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69638211,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 4.024907112121582
    },
    {
      "auxiliary_loss_clip": 0.01169211,
      "auxiliary_loss_mlp": 0.01024436,
      "balance_loss_clip": 0.97229671,
      "balance_loss_mlp": 1.01708031,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 2.217414866191193,
      "language_loss": 0.82758129,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.8495177,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.627129077911377
    },
    {
      "auxiliary_loss_clip": 0.01177238,
      "auxiliary_loss_mlp": 0.0102641,
      "balance_loss_clip": 1.01263785,
      "balance_loss_mlp": 1.01876259,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 1.8081379191263647,
      "language_loss": 0.82967263,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85170913,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.620471477508545
    },
    {
      "auxiliary_loss_clip": 0.0117754,
      "auxiliary_loss_mlp": 0.01027763,
      "balance_loss_clip": 1.05287051,
      "balance_loss_mlp": 1.0204556,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 1.9395327499697337,
      "language_loss": 0.77433288,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.796386,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.6357529163360596
    },
    {
      "auxiliary_loss_clip": 0.01165607,
      "auxiliary_loss_mlp": 0.0112323,
      "balance_loss_clip": 0.89660603,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 2.1611270135219094,
      "language_loss": 0.78136921,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.80425763,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.7675275802612305
    },
    {
      "auxiliary_loss_clip": 0.01167901,
      "auxiliary_loss_mlp": 0.01124087,
      "balance_loss_clip": 0.93728745,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 1.3786218390657268,
      "language_loss": 0.79799509,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.82091498,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.7854630947113037
    },
    {
      "auxiliary_loss_clip": 0.01178502,
      "auxiliary_loss_mlp": 0.01027273,
      "balance_loss_clip": 0.93507564,
      "balance_loss_mlp": 1.01879692,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 1.8150697633336277,
      "language_loss": 0.72034764,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.74240541,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 2.7113404273986816
    },
    {
      "auxiliary_loss_clip": 0.01170953,
      "auxiliary_loss_mlp": 0.01033016,
      "balance_loss_clip": 1.01091254,
      "balance_loss_mlp": 1.02512479,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 3.3626337589589674,
      "language_loss": 0.74650824,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.76854789,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.6774332523345947
    },
    {
      "auxiliary_loss_clip": 0.01181988,
      "auxiliary_loss_mlp": 0.01029154,
      "balance_loss_clip": 0.93911731,
      "balance_loss_mlp": 1.02152205,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 2.1446998255948144,
      "language_loss": 0.77173758,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79384905,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.6814935207366943
    },
    {
      "auxiliary_loss_clip": 0.0107546,
      "auxiliary_loss_mlp": 0.0111682,
      "balance_loss_clip": 0.98059368,
      "balance_loss_mlp": 0.0,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7626167703485714,
      "language_loss": 0.60248744,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.62441027,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.256737232208252
    },
    {
      "auxiliary_loss_clip": 0.01174942,
      "auxiliary_loss_mlp": 0.01027833,
      "balance_loss_clip": 1.01111031,
      "balance_loss_mlp": 1.02028155,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 1.9532909016461715,
      "language_loss": 0.76057088,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.78259867,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 2.6199166774749756
    },
    {
      "auxiliary_loss_clip": 0.01078915,
      "auxiliary_loss_mlp": 0.01000991,
      "balance_loss_clip": 0.94377756,
      "balance_loss_mlp": 0.99890518,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8278212904239187,
      "language_loss": 0.61109257,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63189167,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 3.261406660079956
    },
    {
      "auxiliary_loss_clip": 0.01176033,
      "auxiliary_loss_mlp": 0.01024158,
      "balance_loss_clip": 0.86207432,
      "balance_loss_mlp": 1.01656461,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 2.4759798310822956,
      "language_loss": 0.70261288,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.72461474,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 2.806281566619873
    },
    {
      "auxiliary_loss_clip": 0.01176205,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 0.93906009,
      "balance_loss_mlp": 1.02116275,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 2.0585959840140475,
      "language_loss": 0.77484155,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.79689068,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.7230756282806396
    },
    {
      "auxiliary_loss_clip": 0.01180615,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.05273175,
      "balance_loss_mlp": 1.01778102,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 1.8325477288491352,
      "language_loss": 0.73516572,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.75722808,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 2.6593899726867676
    },
    {
      "auxiliary_loss_clip": 0.01179002,
      "auxiliary_loss_mlp": 0.01031556,
      "balance_loss_clip": 1.01664996,
      "balance_loss_mlp": 1.02426934,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 2.288061185572494,
      "language_loss": 0.78168964,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80379528,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 3.6701674461364746
    },
    {
      "auxiliary_loss_clip": 0.01178996,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 0.97427016,
      "balance_loss_mlp": 1.02214777,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 10.561255078027752,
      "language_loss": 0.82815325,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85023761,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 2.717434883117676
    },
    {
      "auxiliary_loss_clip": 0.01178157,
      "auxiliary_loss_mlp": 0.01122875,
      "balance_loss_clip": 0.97942179,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.608437983281736,
      "language_loss": 0.71001679,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.7330271,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 3.6881227493286133
    },
    {
      "auxiliary_loss_clip": 0.01171415,
      "auxiliary_loss_mlp": 0.01024229,
      "balance_loss_clip": 0.93597639,
      "balance_loss_mlp": 1.01636124,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 1.866714850150994,
      "language_loss": 0.78062475,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80258119,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 2.6905627250671387
    },
    {
      "auxiliary_loss_clip": 0.01159969,
      "auxiliary_loss_mlp": 0.0102586,
      "balance_loss_clip": 0.96977568,
      "balance_loss_mlp": 1.01895237,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 2.055680500562341,
      "language_loss": 0.82856715,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.85042548,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.6393752098083496
    },
    {
      "auxiliary_loss_clip": 0.01169978,
      "auxiliary_loss_mlp": 0.01026211,
      "balance_loss_clip": 0.89830971,
      "balance_loss_mlp": 1.01872444,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 1.695819323071918,
      "language_loss": 0.74132991,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76329184,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.7243921756744385
    },
    {
      "auxiliary_loss_clip": 0.01170449,
      "auxiliary_loss_mlp": 0.01026041,
      "balance_loss_clip": 0.97507203,
      "balance_loss_mlp": 1.0182451,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 1.9591459172573358,
      "language_loss": 0.72664642,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.74861133,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.687453031539917
    },
    {
      "auxiliary_loss_clip": 0.01083595,
      "auxiliary_loss_mlp": 0.01004006,
      "balance_loss_clip": 0.98247862,
      "balance_loss_mlp": 1.00201547,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.8803125212786385,
      "language_loss": 0.60860884,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62948483,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.199812889099121
    },
    {
      "auxiliary_loss_clip": 0.01173261,
      "auxiliary_loss_mlp": 0.01122057,
      "balance_loss_clip": 0.93342417,
      "balance_loss_mlp": 0.0,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.5197649567192624,
      "language_loss": 0.80577302,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.82872623,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 3.751117467880249
    },
    {
      "auxiliary_loss_clip": 0.01182421,
      "auxiliary_loss_mlp": 0.01026059,
      "balance_loss_clip": 1.01434338,
      "balance_loss_mlp": 1.01893604,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 2.056468586707116,
      "language_loss": 0.81560725,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.83769202,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.7709903717041016
    },
    {
      "auxiliary_loss_clip": 0.01170503,
      "auxiliary_loss_mlp": 0.01021499,
      "balance_loss_clip": 0.97452164,
      "balance_loss_mlp": 1.01410794,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 1.6250689160100267,
      "language_loss": 0.73318201,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.75510204,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.945591688156128
    },
    {
      "auxiliary_loss_clip": 0.01177611,
      "auxiliary_loss_mlp": 0.01123099,
      "balance_loss_clip": 1.01427078,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 1.8208350959062822,
      "language_loss": 0.82549411,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84850121,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 3.6596832275390625
    },
    {
      "auxiliary_loss_clip": 0.01167425,
      "auxiliary_loss_mlp": 0.01022005,
      "balance_loss_clip": 0.93561602,
      "balance_loss_mlp": 1.01460767,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 2.7400016645109693,
      "language_loss": 0.8025009,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.82439518,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.7231571674346924
    },
    {
      "auxiliary_loss_clip": 0.01179782,
      "auxiliary_loss_mlp": 0.0103534,
      "balance_loss_clip": 0.97952199,
      "balance_loss_mlp": 1.02808642,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 1.9880608698593258,
      "language_loss": 0.80141687,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.82356805,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.705195188522339
    },
    {
      "auxiliary_loss_clip": 0.01169548,
      "auxiliary_loss_mlp": 0.01031824,
      "balance_loss_clip": 0.97471237,
      "balance_loss_mlp": 1.02394998,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.133217309600125,
      "language_loss": 0.66350913,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68552279,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.6863112449645996
    },
    {
      "auxiliary_loss_clip": 0.01183265,
      "auxiliary_loss_mlp": 0.01024123,
      "balance_loss_clip": 1.01424694,
      "balance_loss_mlp": 1.01707196,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 1.7439656242403034,
      "language_loss": 0.7859894,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.80806327,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.661059617996216
    },
    {
      "auxiliary_loss_clip": 0.01172278,
      "auxiliary_loss_mlp": 0.01027569,
      "balance_loss_clip": 0.97451741,
      "balance_loss_mlp": 1.01973081,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 2.465164586967833,
      "language_loss": 0.67257142,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.69456995,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.6825106143951416
    },
    {
      "auxiliary_loss_clip": 0.01176644,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 0.97485167,
      "balance_loss_mlp": 1.02074337,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 1.8303689111761716,
      "language_loss": 0.63614428,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.65819776,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 2.697883367538452
    },
    {
      "auxiliary_loss_clip": 0.01180575,
      "auxiliary_loss_mlp": 0.01024471,
      "balance_loss_clip": 1.0533911,
      "balance_loss_mlp": 1.01673388,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 1.8814606006180679,
      "language_loss": 0.72102082,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74307132,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.6364872455596924
    },
    {
      "auxiliary_loss_clip": 0.01166977,
      "auxiliary_loss_mlp": 0.01023576,
      "balance_loss_clip": 0.93766433,
      "balance_loss_mlp": 1.01618171,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 1.535921755359754,
      "language_loss": 0.78114569,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80305123,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.720353603363037
    },
    {
      "auxiliary_loss_clip": 0.01165394,
      "auxiliary_loss_mlp": 0.01023794,
      "balance_loss_clip": 0.85252005,
      "balance_loss_mlp": 1.01591992,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 2.4190460762000674,
      "language_loss": 0.71754372,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.73943561,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.813106060028076
    },
    {
      "auxiliary_loss_clip": 0.01178317,
      "auxiliary_loss_mlp": 0.01032019,
      "balance_loss_clip": 0.93600821,
      "balance_loss_mlp": 1.02432752,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.2627551687327196,
      "language_loss": 0.83008909,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85219252,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.6845619678497314
    },
    {
      "auxiliary_loss_clip": 0.01178017,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.04959822,
      "balance_loss_mlp": 1.01781094,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 2.3235905034182767,
      "language_loss": 0.81539559,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.83742994,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 2.629756212234497
    },
    {
      "auxiliary_loss_clip": 0.01184147,
      "auxiliary_loss_mlp": 0.01028452,
      "balance_loss_clip": 1.01595306,
      "balance_loss_mlp": 1.02158213,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.2179310529822347,
      "language_loss": 0.75710618,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.77923214,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 2.716343879699707
    },
    {
      "auxiliary_loss_clip": 0.01168248,
      "auxiliary_loss_mlp": 0.01023563,
      "balance_loss_clip": 0.8961091,
      "balance_loss_mlp": 1.0162499,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 2.4677711213905855,
      "language_loss": 0.78325349,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.80517161,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.6764845848083496
    },
    {
      "auxiliary_loss_clip": 0.01170053,
      "auxiliary_loss_mlp": 0.0102736,
      "balance_loss_clip": 0.93409497,
      "balance_loss_mlp": 1.01943874,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 2.211025706794542,
      "language_loss": 0.79403526,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.8160094,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 2.7432777881622314
    },
    {
      "auxiliary_loss_clip": 0.01175166,
      "auxiliary_loss_mlp": 0.01031847,
      "balance_loss_clip": 0.89851093,
      "balance_loss_mlp": 1.02453995,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.0602174655568244,
      "language_loss": 0.72602206,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74809217,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 3.5961263179779053
    },
    {
      "auxiliary_loss_clip": 0.01177287,
      "auxiliary_loss_mlp": 0.01024503,
      "balance_loss_clip": 1.05309772,
      "balance_loss_mlp": 1.01783013,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 1.5761524060851784,
      "language_loss": 0.75276625,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.77478415,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.7959485054016113
    },
    {
      "auxiliary_loss_clip": 0.01184275,
      "auxiliary_loss_mlp": 0.01024539,
      "balance_loss_clip": 0.93500876,
      "balance_loss_mlp": 1.01736879,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 1.8160179121030966,
      "language_loss": 0.78136063,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.8034488,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 3.691298007965088
    },
    {
      "auxiliary_loss_clip": 0.01173747,
      "auxiliary_loss_mlp": 0.01029134,
      "balance_loss_clip": 0.93965954,
      "balance_loss_mlp": 1.02201724,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 1.6977247451686124,
      "language_loss": 0.8289364,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.8509652,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 2.7014856338500977
    },
    {
      "auxiliary_loss_clip": 0.01172637,
      "auxiliary_loss_mlp": 0.01020766,
      "balance_loss_clip": 1.01316905,
      "balance_loss_mlp": 1.01363182,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 1.6578756705229662,
      "language_loss": 0.71789753,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.73983157,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.7153923511505127
    },
    {
      "auxiliary_loss_clip": 0.01174154,
      "auxiliary_loss_mlp": 0.01122534,
      "balance_loss_clip": 0.97732353,
      "balance_loss_mlp": 0.0,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 2.1116124302985497,
      "language_loss": 0.77902985,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.80199671,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.737135887145996
    },
    {
      "auxiliary_loss_clip": 0.01177993,
      "auxiliary_loss_mlp": 0.01030126,
      "balance_loss_clip": 0.9768551,
      "balance_loss_mlp": 1.02312875,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 1.7345832957250824,
      "language_loss": 0.75704098,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77912223,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.6541507244110107
    },
    {
      "auxiliary_loss_clip": 0.01184204,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 0.97530526,
      "balance_loss_mlp": 1.022403,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 2.188588170586712,
      "language_loss": 0.77275717,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.79490012,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.790452480316162
    },
    {
      "auxiliary_loss_clip": 0.01189251,
      "auxiliary_loss_mlp": 0.01028448,
      "balance_loss_clip": 0.97654134,
      "balance_loss_mlp": 1.02033591,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 2.598407428554456,
      "language_loss": 0.6771276,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.69930452,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 3.6389999389648438
    },
    {
      "auxiliary_loss_clip": 0.01165732,
      "auxiliary_loss_mlp": 0.01122786,
      "balance_loss_clip": 0.93289781,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 3.0894267235476978,
      "language_loss": 0.78692985,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80981499,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.7622578144073486
    },
    {
      "auxiliary_loss_clip": 0.01099124,
      "auxiliary_loss_mlp": 0.01006266,
      "balance_loss_clip": 0.96511149,
      "balance_loss_mlp": 1.00432265,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.6948341948732175,
      "language_loss": 0.54187119,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56292498,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.367311954498291
    },
    {
      "auxiliary_loss_clip": 0.01171772,
      "auxiliary_loss_mlp": 0.01023286,
      "balance_loss_clip": 1.01281607,
      "balance_loss_mlp": 1.01667273,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 1.6741964411632941,
      "language_loss": 0.77637756,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79832816,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 3.5254809856414795
    },
    {
      "auxiliary_loss_clip": 0.01169362,
      "auxiliary_loss_mlp": 0.01028449,
      "balance_loss_clip": 0.9348076,
      "balance_loss_mlp": 1.02110577,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 2.5051023205555665,
      "language_loss": 0.76198971,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.78396785,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.7636117935180664
    },
    {
      "auxiliary_loss_clip": 0.01177846,
      "auxiliary_loss_mlp": 0.01031099,
      "balance_loss_clip": 0.97662103,
      "balance_loss_mlp": 1.02305269,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 2.293068687379989,
      "language_loss": 0.67708445,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.69917393,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.7035205364227295
    },
    {
      "auxiliary_loss_clip": 0.01180616,
      "auxiliary_loss_mlp": 0.01030733,
      "balance_loss_clip": 1.05453265,
      "balance_loss_mlp": 1.02409875,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.7357478232558492,
      "language_loss": 0.74671477,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76882827,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.7118144035339355
    },
    {
      "auxiliary_loss_clip": 0.01165549,
      "auxiliary_loss_mlp": 0.01024643,
      "balance_loss_clip": 0.97278237,
      "balance_loss_mlp": 1.01764584,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 1.6409087897339267,
      "language_loss": 0.81914282,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.84104472,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.7602739334106445
    },
    {
      "auxiliary_loss_clip": 0.0117407,
      "auxiliary_loss_mlp": 0.01030077,
      "balance_loss_clip": 1.01030779,
      "balance_loss_mlp": 1.02260232,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.6366350873230442,
      "language_loss": 0.83919722,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.86123872,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.6844820976257324
    },
    {
      "auxiliary_loss_clip": 0.0117574,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 0.97218776,
      "balance_loss_mlp": 1.02095497,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 1.669388118659948,
      "language_loss": 0.79047465,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.81252193,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 2.7018470764160156
    },
    {
      "auxiliary_loss_clip": 0.01175523,
      "auxiliary_loss_mlp": 0.01026799,
      "balance_loss_clip": 0.97452915,
      "balance_loss_mlp": 1.01964378,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 2.53788125988837,
      "language_loss": 0.69605047,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.71807373,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.7441184520721436
    },
    {
      "auxiliary_loss_clip": 0.01173391,
      "auxiliary_loss_mlp": 0.01024947,
      "balance_loss_clip": 0.97455889,
      "balance_loss_mlp": 1.01778543,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.704185440983328,
      "language_loss": 0.7948457,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81682903,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.7231593132019043
    },
    {
      "auxiliary_loss_clip": 0.01178587,
      "auxiliary_loss_mlp": 0.01024745,
      "balance_loss_clip": 1.0133357,
      "balance_loss_mlp": 1.01681161,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 1.7020689835463125,
      "language_loss": 0.79802322,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.8200565,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.702937126159668
    },
    {
      "auxiliary_loss_clip": 0.0108504,
      "auxiliary_loss_mlp": 0.01004888,
      "balance_loss_clip": 0.94860893,
      "balance_loss_mlp": 1.00298107,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.8578407566624827,
      "language_loss": 0.52876908,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.54966837,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.1705968379974365
    },
    {
      "auxiliary_loss_clip": 0.01163062,
      "auxiliary_loss_mlp": 0.01029463,
      "balance_loss_clip": 0.85618615,
      "balance_loss_mlp": 1.02151215,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 1.9664460370500743,
      "language_loss": 0.75632131,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.77824652,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 2.780057430267334
    },
    {
      "auxiliary_loss_clip": 0.01180401,
      "auxiliary_loss_mlp": 0.01027094,
      "balance_loss_clip": 0.93943202,
      "balance_loss_mlp": 1.01981068,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 1.7568787549697686,
      "language_loss": 0.87735403,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.89942896,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 2.7103142738342285
    },
    {
      "auxiliary_loss_clip": 0.01179076,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 1.01227117,
      "balance_loss_mlp": 1.01886702,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 1.7698151706199332,
      "language_loss": 0.64878428,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67084599,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 2.671438694000244
    },
    {
      "auxiliary_loss_clip": 0.01176848,
      "auxiliary_loss_mlp": 0.01025202,
      "balance_loss_clip": 1.05374169,
      "balance_loss_mlp": 1.01864839,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 1.7685387770330065,
      "language_loss": 0.77724177,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.79926229,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 2.66357421875
    },
    {
      "auxiliary_loss_clip": 0.01186329,
      "auxiliary_loss_mlp": 0.01123231,
      "balance_loss_clip": 1.0165689,
      "balance_loss_mlp": 0.0,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 2.1020164130370596,
      "language_loss": 0.80192745,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.82502306,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 3.681401014328003
    },
    {
      "auxiliary_loss_clip": 0.01184528,
      "auxiliary_loss_mlp": 0.01034433,
      "balance_loss_clip": 1.05338025,
      "balance_loss_mlp": 1.02640438,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 2.8174572028640466,
      "language_loss": 0.62257099,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.64476061,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 2.6149163246154785
    },
    {
      "auxiliary_loss_clip": 0.01167352,
      "auxiliary_loss_mlp": 0.01029563,
      "balance_loss_clip": 0.93807679,
      "balance_loss_mlp": 1.02175522,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 1.6799859359212117,
      "language_loss": 0.64945942,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.67142856,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 3.75087571144104
    },
    {
      "auxiliary_loss_clip": 0.01177203,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 0.93501878,
      "balance_loss_mlp": 1.01735663,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 1.7382807567523708,
      "language_loss": 0.72118533,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.7432009,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.8333818912506104
    },
    {
      "auxiliary_loss_clip": 0.0118087,
      "auxiliary_loss_mlp": 0.01029454,
      "balance_loss_clip": 1.05163312,
      "balance_loss_mlp": 1.02138638,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 1.833084383224113,
      "language_loss": 0.8195194,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84162271,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.688155174255371
    },
    {
      "auxiliary_loss_clip": 0.01185381,
      "auxiliary_loss_mlp": 0.01025521,
      "balance_loss_clip": 1.0150367,
      "balance_loss_mlp": 1.01769781,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.0915574712156695,
      "language_loss": 0.79695129,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.81906033,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.5875747203826904
    },
    {
      "auxiliary_loss_clip": 0.01165082,
      "auxiliary_loss_mlp": 0.01025052,
      "balance_loss_clip": 1.00968742,
      "balance_loss_mlp": 1.01779532,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.7923986823201021,
      "language_loss": 0.73428202,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75618333,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.746382236480713
    },
    {
      "auxiliary_loss_clip": 0.0118036,
      "auxiliary_loss_mlp": 0.01122085,
      "balance_loss_clip": 1.01511717,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 1.8053113492790507,
      "language_loss": 0.67302895,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69605339,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.687880277633667
    },
    {
      "auxiliary_loss_clip": 0.01172986,
      "auxiliary_loss_mlp": 0.01026318,
      "balance_loss_clip": 0.93659455,
      "balance_loss_mlp": 1.01927304,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 2.1227403509933644,
      "language_loss": 0.72857684,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.75056982,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 3.6499698162078857
    },
    {
      "auxiliary_loss_clip": 0.01175164,
      "auxiliary_loss_mlp": 0.01022276,
      "balance_loss_clip": 0.90012026,
      "balance_loss_mlp": 1.0147717,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 1.835100584348888,
      "language_loss": 0.79209381,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.8140682,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.7676455974578857
    },
    {
      "auxiliary_loss_clip": 0.01076154,
      "auxiliary_loss_mlp": 0.01003147,
      "balance_loss_clip": 0.94464302,
      "balance_loss_mlp": 1.0011797,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 2.9758619597275904,
      "language_loss": 0.57388514,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.59467816,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.1479737758636475
    },
    {
      "auxiliary_loss_clip": 0.01077697,
      "auxiliary_loss_mlp": 0.01004048,
      "balance_loss_clip": 0.86754,
      "balance_loss_mlp": 1.00200927,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8713234286235637,
      "language_loss": 0.62115085,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.64196825,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 4.407903671264648
    },
    {
      "auxiliary_loss_clip": 0.01173621,
      "auxiliary_loss_mlp": 0.01122903,
      "balance_loss_clip": 0.93605119,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 2.9950573693115863,
      "language_loss": 0.84375143,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.86671662,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.7348744869232178
    },
    {
      "auxiliary_loss_clip": 0.01181556,
      "auxiliary_loss_mlp": 0.01028286,
      "balance_loss_clip": 1.01414275,
      "balance_loss_mlp": 1.0211215,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 1.7944185554622516,
      "language_loss": 0.85923249,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.88133091,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.7251136302948
    },
    {
      "auxiliary_loss_clip": 0.0117054,
      "auxiliary_loss_mlp": 0.0112263,
      "balance_loss_clip": 0.93724895,
      "balance_loss_mlp": 0.0,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 1.7870995412912334,
      "language_loss": 0.79303503,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.81596678,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.695770502090454
    },
    {
      "auxiliary_loss_clip": 0.01160474,
      "auxiliary_loss_mlp": 0.01029649,
      "balance_loss_clip": 0.93736291,
      "balance_loss_mlp": 1.02272928,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 1.6769631765513042,
      "language_loss": 0.82412612,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.84602737,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.8294429779052734
    },
    {
      "auxiliary_loss_clip": 0.01176896,
      "auxiliary_loss_mlp": 0.01028121,
      "balance_loss_clip": 1.01135325,
      "balance_loss_mlp": 1.02092385,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 4.169659251276357,
      "language_loss": 0.78796774,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.81001788,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.711298942565918
    },
    {
      "auxiliary_loss_clip": 0.01177403,
      "auxiliary_loss_mlp": 0.01029486,
      "balance_loss_clip": 0.97414935,
      "balance_loss_mlp": 1.02146864,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 3.921976603758382,
      "language_loss": 0.67566353,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.69773233,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 2.695185661315918
    },
    {
      "auxiliary_loss_clip": 0.01178722,
      "auxiliary_loss_mlp": 0.01032838,
      "balance_loss_clip": 1.01578367,
      "balance_loss_mlp": 1.02544975,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.1490412206190346,
      "language_loss": 0.79189432,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.8140099,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.62819504737854
    },
    {
      "auxiliary_loss_clip": 0.01170814,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 0.90070623,
      "balance_loss_mlp": 1.02054834,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 1.8170428841689426,
      "language_loss": 0.73313105,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.75512022,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.7863519191741943
    },
    {
      "auxiliary_loss_clip": 0.01183216,
      "auxiliary_loss_mlp": 0.01123272,
      "balance_loss_clip": 0.93846136,
      "balance_loss_mlp": 0.0,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 2.024084044596419,
      "language_loss": 0.72151846,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.74458325,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.692044973373413
    },
    {
      "auxiliary_loss_clip": 0.01172673,
      "auxiliary_loss_mlp": 0.01030279,
      "balance_loss_clip": 0.93797731,
      "balance_loss_mlp": 1.02192807,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 1.8111693572151453,
      "language_loss": 0.69989121,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72192073,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 2.681426763534546
    },
    {
      "auxiliary_loss_clip": 0.01175247,
      "auxiliary_loss_mlp": 0.0102885,
      "balance_loss_clip": 0.97119975,
      "balance_loss_mlp": 1.02122653,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 1.9500244876283281,
      "language_loss": 0.81498688,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.83702779,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 2.6638495922088623
    },
    {
      "auxiliary_loss_clip": 0.01178351,
      "auxiliary_loss_mlp": 0.01026546,
      "balance_loss_clip": 0.97453475,
      "balance_loss_mlp": 1.01919699,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 1.706102932916757,
      "language_loss": 0.74479645,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.76684546,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 2.6504688262939453
    },
    {
      "auxiliary_loss_clip": 0.01176276,
      "auxiliary_loss_mlp": 0.01031185,
      "balance_loss_clip": 0.97515357,
      "balance_loss_mlp": 1.02366614,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 1.810650029354026,
      "language_loss": 0.86628914,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.88836372,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 2.6394169330596924
    },
    {
      "auxiliary_loss_clip": 0.01177982,
      "auxiliary_loss_mlp": 0.01030013,
      "balance_loss_clip": 0.9756906,
      "balance_loss_mlp": 1.02186775,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 1.7939991747894781,
      "language_loss": 0.66983622,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69191617,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 3.6817121505737305
    },
    {
      "auxiliary_loss_clip": 0.01086387,
      "auxiliary_loss_mlp": 0.0100856,
      "balance_loss_clip": 0.94613194,
      "balance_loss_mlp": 1.00653315,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.0788034113967944,
      "language_loss": 0.77384454,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79479402,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.322200298309326
    },
    {
      "auxiliary_loss_clip": 0.01183744,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 0.97458893,
      "balance_loss_mlp": 1.01789045,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 1.8284727290343255,
      "language_loss": 0.90627992,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.92837048,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.7034919261932373
    },
    {
      "auxiliary_loss_clip": 0.01180644,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 0.90005165,
      "balance_loss_mlp": 1.01532364,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 1.5443378297739403,
      "language_loss": 0.79323387,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.8152709,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 3.6880617141723633
    },
    {
      "auxiliary_loss_clip": 0.01173616,
      "auxiliary_loss_mlp": 0.01029096,
      "balance_loss_clip": 0.97220755,
      "balance_loss_mlp": 1.02103186,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 1.6433135580675922,
      "language_loss": 0.76872802,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.79075515,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 2.725044012069702
    },
    {
      "auxiliary_loss_clip": 0.01180051,
      "auxiliary_loss_mlp": 0.01027366,
      "balance_loss_clip": 1.01669729,
      "balance_loss_mlp": 1.02020502,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 2.0091504664354694,
      "language_loss": 0.80660498,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.8286792,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.702565908432007
    },
    {
      "auxiliary_loss_clip": 0.01180742,
      "auxiliary_loss_mlp": 0.01031357,
      "balance_loss_clip": 0.93803841,
      "balance_loss_mlp": 1.0238167,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.550561242067185,
      "language_loss": 0.82838243,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.85050344,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.838564395904541
    },
    {
      "auxiliary_loss_clip": 0.01179086,
      "auxiliary_loss_mlp": 0.01029155,
      "balance_loss_clip": 1.01483011,
      "balance_loss_mlp": 1.02160311,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 1.6599918066635189,
      "language_loss": 0.69606459,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.71814698,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 2.5829339027404785
    },
    {
      "auxiliary_loss_clip": 0.01177406,
      "auxiliary_loss_mlp": 0.01035604,
      "balance_loss_clip": 0.89540678,
      "balance_loss_mlp": 1.02754557,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 2.4637420817238307,
      "language_loss": 0.70736933,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.72949946,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.7323105335235596
    },
    {
      "auxiliary_loss_clip": 0.01175661,
      "auxiliary_loss_mlp": 0.01028535,
      "balance_loss_clip": 0.97579348,
      "balance_loss_mlp": 1.02088165,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 2.2619532689119826,
      "language_loss": 0.78068161,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80272359,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 3.523442506790161
    },
    {
      "auxiliary_loss_clip": 0.01176148,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 0.93523955,
      "balance_loss_mlp": 1.01788378,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 1.8489084781298366,
      "language_loss": 0.67840952,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.70042574,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.6992571353912354
    },
    {
      "auxiliary_loss_clip": 0.01180058,
      "auxiliary_loss_mlp": 0.01123685,
      "balance_loss_clip": 1.01292539,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 1.7994251593074184,
      "language_loss": 0.7566669,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77970433,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 3.718390464782715
    },
    {
      "auxiliary_loss_clip": 0.01181461,
      "auxiliary_loss_mlp": 0.01123295,
      "balance_loss_clip": 0.97536004,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 1.654216566440082,
      "language_loss": 0.78771937,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.81076688,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.71895170211792
    },
    {
      "auxiliary_loss_clip": 0.01172188,
      "auxiliary_loss_mlp": 0.01025646,
      "balance_loss_clip": 0.97494078,
      "balance_loss_mlp": 1.01846993,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 1.7416115428455472,
      "language_loss": 0.84166002,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.8636384,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.7351791858673096
    },
    {
      "auxiliary_loss_clip": 0.01179563,
      "auxiliary_loss_mlp": 0.01023212,
      "balance_loss_clip": 1.04993367,
      "balance_loss_mlp": 1.01576138,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 1.6117591840773973,
      "language_loss": 0.76710868,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.78913641,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.715363025665283
    },
    {
      "auxiliary_loss_clip": 0.01180134,
      "auxiliary_loss_mlp": 0.01024684,
      "balance_loss_clip": 1.0500952,
      "balance_loss_mlp": 1.01655996,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 1.6633563710835801,
      "language_loss": 0.81680763,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.83885574,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.7305288314819336
    },
    {
      "auxiliary_loss_clip": 0.0116861,
      "auxiliary_loss_mlp": 0.01123516,
      "balance_loss_clip": 0.97380221,
      "balance_loss_mlp": 0.0,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 2.1960005595245313,
      "language_loss": 0.61744988,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.64037114,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.6856791973114014
    },
    {
      "auxiliary_loss_clip": 0.01183446,
      "auxiliary_loss_mlp": 0.01022976,
      "balance_loss_clip": 1.01437974,
      "balance_loss_mlp": 1.01594234,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 1.7387357104759882,
      "language_loss": 0.72327417,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.74533838,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.725222587585449
    },
    {
      "auxiliary_loss_clip": 0.01179468,
      "auxiliary_loss_mlp": 0.01031245,
      "balance_loss_clip": 1.01427579,
      "balance_loss_mlp": 1.0239495,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 1.967269786994216,
      "language_loss": 0.69241154,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71451867,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 2.6666338443756104
    },
    {
      "auxiliary_loss_clip": 0.0118294,
      "auxiliary_loss_mlp": 0.01032672,
      "balance_loss_clip": 0.97780788,
      "balance_loss_mlp": 1.0248816,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 1.8038495720885104,
      "language_loss": 0.70874161,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73089772,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.700735569000244
    },
    {
      "auxiliary_loss_clip": 0.01170314,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 0.97271514,
      "balance_loss_mlp": 1.01901865,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 1.5652639495022476,
      "language_loss": 0.80982691,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.83179784,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.706477642059326
    },
    {
      "auxiliary_loss_clip": 0.01182568,
      "auxiliary_loss_mlp": 0.01023646,
      "balance_loss_clip": 0.93982464,
      "balance_loss_mlp": 1.01669645,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 2.414597760079881,
      "language_loss": 0.81054604,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.83260822,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.744919538497925
    },
    {
      "auxiliary_loss_clip": 0.01172857,
      "auxiliary_loss_mlp": 0.01025277,
      "balance_loss_clip": 0.97301501,
      "balance_loss_mlp": 1.01793706,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 1.7569558062216275,
      "language_loss": 0.74622887,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76821017,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 2.74008846282959
    },
    {
      "auxiliary_loss_clip": 0.0117964,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 1.01466656,
      "balance_loss_mlp": 1.02099323,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 2.058855951545856,
      "language_loss": 0.72413415,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74622309,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 2.646038055419922
    },
    {
      "auxiliary_loss_clip": 0.01175595,
      "auxiliary_loss_mlp": 0.01027213,
      "balance_loss_clip": 0.93768156,
      "balance_loss_mlp": 1.01975632,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.3003359298080093,
      "language_loss": 0.86167061,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.8836987,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 2.6809465885162354
    },
    {
      "auxiliary_loss_clip": 0.01181572,
      "auxiliary_loss_mlp": 0.0102402,
      "balance_loss_clip": 0.97589767,
      "balance_loss_mlp": 1.01604486,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 1.9264163014797409,
      "language_loss": 0.77269387,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.79474986,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 2.7422497272491455
    },
    {
      "auxiliary_loss_clip": 0.01183391,
      "auxiliary_loss_mlp": 0.01031902,
      "balance_loss_clip": 1.05336642,
      "balance_loss_mlp": 1.02418351,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 1.797263128441114,
      "language_loss": 0.71853733,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.74069029,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 3.6253254413604736
    },
    {
      "auxiliary_loss_clip": 0.0118199,
      "auxiliary_loss_mlp": 0.01027491,
      "balance_loss_clip": 1.01305604,
      "balance_loss_mlp": 1.01960576,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 3.9782929331321064,
      "language_loss": 0.87285912,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89495397,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.63743257522583
    },
    {
      "auxiliary_loss_clip": 0.01180642,
      "auxiliary_loss_mlp": 0.01025713,
      "balance_loss_clip": 0.97608185,
      "balance_loss_mlp": 1.01862025,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.3714335163068392,
      "language_loss": 0.76507777,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.78714132,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 2.6972169876098633
    },
    {
      "auxiliary_loss_clip": 0.01182398,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.01774943,
      "balance_loss_mlp": 1.02384472,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.6834618663330323,
      "language_loss": 0.87006915,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.89220566,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 3.531752109527588
    },
    {
      "auxiliary_loss_clip": 0.01177849,
      "auxiliary_loss_mlp": 0.01029415,
      "balance_loss_clip": 1.05236065,
      "balance_loss_mlp": 1.02232814,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 2.170169084522847,
      "language_loss": 0.71705788,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.7391305,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 2.7342898845672607
    },
    {
      "auxiliary_loss_clip": 0.01179287,
      "auxiliary_loss_mlp": 0.01024261,
      "balance_loss_clip": 1.05204582,
      "balance_loss_mlp": 1.0165658,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 2.0954758986827606,
      "language_loss": 0.72839618,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.75043166,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.6962249279022217
    },
    {
      "auxiliary_loss_clip": 0.01187584,
      "auxiliary_loss_mlp": 0.01027783,
      "balance_loss_clip": 0.90743238,
      "balance_loss_mlp": 1.0204041,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 1.7558206412402686,
      "language_loss": 0.64885092,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.67100459,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.7259726524353027
    },
    {
      "auxiliary_loss_clip": 0.01175512,
      "auxiliary_loss_mlp": 0.01031418,
      "balance_loss_clip": 1.01399577,
      "balance_loss_mlp": 1.02364552,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 1.7281584559455256,
      "language_loss": 0.75729942,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.7793687,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.705813407897949
    },
    {
      "auxiliary_loss_clip": 0.01180394,
      "auxiliary_loss_mlp": 0.01036075,
      "balance_loss_clip": 1.05308938,
      "balance_loss_mlp": 1.02853489,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 2.325343181957132,
      "language_loss": 0.79863012,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.82079482,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.6147818565368652
    },
    {
      "auxiliary_loss_clip": 0.01156758,
      "auxiliary_loss_mlp": 0.01121864,
      "balance_loss_clip": 0.89529216,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 3.102778973575557,
      "language_loss": 0.73973465,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.76252091,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 3.9596378803253174
    },
    {
      "auxiliary_loss_clip": 0.01181521,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 0.97765517,
      "balance_loss_mlp": 1.02380347,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 2.477954275315522,
      "language_loss": 0.80404711,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.82618171,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.675793409347534
    },
    {
      "auxiliary_loss_clip": 0.01177086,
      "auxiliary_loss_mlp": 0.01025058,
      "balance_loss_clip": 0.97423786,
      "balance_loss_mlp": 1.01695228,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 1.8575167206275922,
      "language_loss": 0.75098264,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77300406,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.6192283630371094
    },
    {
      "auxiliary_loss_clip": 0.01092657,
      "auxiliary_loss_mlp": 0.01116997,
      "balance_loss_clip": 0.87042785,
      "balance_loss_mlp": 0.0,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.8912672425675533,
      "language_loss": 0.63150692,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.65360343,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 4.307246923446655
    },
    {
      "auxiliary_loss_clip": 0.01177445,
      "auxiliary_loss_mlp": 0.01033076,
      "balance_loss_clip": 1.05164099,
      "balance_loss_mlp": 1.02566099,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 1.5711564710819106,
      "language_loss": 0.71032643,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.73243165,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.8226418495178223
    },
    {
      "auxiliary_loss_clip": 0.01176544,
      "auxiliary_loss_mlp": 0.01027581,
      "balance_loss_clip": 0.97497636,
      "balance_loss_mlp": 1.02011538,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 1.7912982564051783,
      "language_loss": 0.80378288,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82582414,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.718397855758667
    },
    {
      "auxiliary_loss_clip": 0.0117505,
      "auxiliary_loss_mlp": 0.0103264,
      "balance_loss_clip": 0.89813274,
      "balance_loss_mlp": 1.02508783,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 2.489597605711961,
      "language_loss": 0.82763702,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.84971392,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.7061989307403564
    },
    {
      "auxiliary_loss_clip": 0.01173582,
      "auxiliary_loss_mlp": 0.01028555,
      "balance_loss_clip": 0.97436982,
      "balance_loss_mlp": 1.02078283,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.160508054653871,
      "language_loss": 0.76316309,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.78518444,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.624565362930298
    },
    {
      "auxiliary_loss_clip": 0.01178881,
      "auxiliary_loss_mlp": 0.01121831,
      "balance_loss_clip": 1.05534708,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.8352243677991764,
      "language_loss": 0.88424087,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90724802,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.724445343017578
    },
    {
      "auxiliary_loss_clip": 0.01166195,
      "auxiliary_loss_mlp": 0.01025258,
      "balance_loss_clip": 0.97352552,
      "balance_loss_mlp": 1.01844764,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 1.7693289045704352,
      "language_loss": 0.78110313,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80301762,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 2.7594215869903564
    },
    {
      "auxiliary_loss_clip": 0.01185789,
      "auxiliary_loss_mlp": 0.01029244,
      "balance_loss_clip": 1.01631117,
      "balance_loss_mlp": 1.02094698,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 3.320263752628049,
      "language_loss": 0.68776071,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.70991099,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.72031831741333
    },
    {
      "auxiliary_loss_clip": 0.01173664,
      "auxiliary_loss_mlp": 0.01028329,
      "balance_loss_clip": 1.01037049,
      "balance_loss_mlp": 1.02119994,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.3508810162305884,
      "language_loss": 0.79053283,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.81255275,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.763679265975952
    },
    {
      "auxiliary_loss_clip": 0.01086835,
      "auxiliary_loss_mlp": 0.01005708,
      "balance_loss_clip": 0.90932262,
      "balance_loss_mlp": 1.00368118,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.866631903298788,
      "language_loss": 0.59953338,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.62045884,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.306649684906006
    },
    {
      "auxiliary_loss_clip": 0.01176606,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 0.93433213,
      "balance_loss_mlp": 1.01761472,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.090862946220142,
      "language_loss": 0.8245014,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.84651601,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 2.727210760116577
    },
    {
      "auxiliary_loss_clip": 0.01175998,
      "auxiliary_loss_mlp": 0.01030008,
      "balance_loss_clip": 1.0124203,
      "balance_loss_mlp": 1.02196097,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 1.6905505991557361,
      "language_loss": 0.82440054,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84646058,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 2.64514422416687
    },
    {
      "auxiliary_loss_clip": 0.01172818,
      "auxiliary_loss_mlp": 0.01026844,
      "balance_loss_clip": 0.93471885,
      "balance_loss_mlp": 1.01910734,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 1.658544030487855,
      "language_loss": 0.70528156,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72727823,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 2.769702434539795
    },
    {
      "auxiliary_loss_clip": 0.01174291,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 0.89810348,
      "balance_loss_mlp": 1.01594198,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.5003297946590433,
      "language_loss": 0.710199,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.73217845,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 2.7332167625427246
    },
    {
      "auxiliary_loss_clip": 0.01174098,
      "auxiliary_loss_mlp": 0.01025569,
      "balance_loss_clip": 1.01177526,
      "balance_loss_mlp": 1.01790369,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 1.591231679391238,
      "language_loss": 0.74111497,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.76311159,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 3.550297975540161
    },
    {
      "auxiliary_loss_clip": 0.01163969,
      "auxiliary_loss_mlp": 0.01029615,
      "balance_loss_clip": 0.89402604,
      "balance_loss_mlp": 1.02145505,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 1.6368246149786154,
      "language_loss": 0.72897792,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75091374,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.7844996452331543
    },
    {
      "auxiliary_loss_clip": 0.01179179,
      "auxiliary_loss_mlp": 0.01032146,
      "balance_loss_clip": 0.9712503,
      "balance_loss_mlp": 1.02437973,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 1.6302873124588853,
      "language_loss": 0.87081981,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89293301,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.6788792610168457
    },
    {
      "auxiliary_loss_clip": 0.01169873,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 0.89771974,
      "balance_loss_mlp": 1.02003241,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 1.964139384514205,
      "language_loss": 0.58118278,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60316193,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 3.7047109603881836
    },
    {
      "auxiliary_loss_clip": 0.01175458,
      "auxiliary_loss_mlp": 0.01021983,
      "balance_loss_clip": 0.93446672,
      "balance_loss_mlp": 1.01435935,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 1.9984236521948644,
      "language_loss": 0.82596302,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.84793741,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 2.6650378704071045
    },
    {
      "auxiliary_loss_clip": 0.01178628,
      "auxiliary_loss_mlp": 0.01027957,
      "balance_loss_clip": 0.90091145,
      "balance_loss_mlp": 1.01968384,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 2.3617349655748785,
      "language_loss": 0.79222119,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.81428707,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.737642526626587
    },
    {
      "auxiliary_loss_clip": 0.0118641,
      "auxiliary_loss_mlp": 0.01034924,
      "balance_loss_clip": 1.05686617,
      "balance_loss_mlp": 1.02680027,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 3.153422930863971,
      "language_loss": 0.84756196,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.8697753,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.6156067848205566
    },
    {
      "auxiliary_loss_clip": 0.01185189,
      "auxiliary_loss_mlp": 0.0102654,
      "balance_loss_clip": 1.01386976,
      "balance_loss_mlp": 1.01825511,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 2.366339799250279,
      "language_loss": 0.73675108,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.75886834,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 2.734835386276245
    },
    {
      "auxiliary_loss_clip": 0.01167131,
      "auxiliary_loss_mlp": 0.0102764,
      "balance_loss_clip": 0.89684379,
      "balance_loss_mlp": 1.01983738,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 1.8270820326590975,
      "language_loss": 0.74702102,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.7689687,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 3.6757805347442627
    },
    {
      "auxiliary_loss_clip": 0.01084972,
      "auxiliary_loss_mlp": 0.01002203,
      "balance_loss_clip": 0.98467147,
      "balance_loss_mlp": 1.00014043,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8544144578041398,
      "language_loss": 0.6198045,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64067626,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.356351375579834
    },
    {
      "auxiliary_loss_clip": 0.01180314,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.05203652,
      "balance_loss_mlp": 1.01966274,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 2.0620733305414958,
      "language_loss": 0.73813343,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.76020789,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.6012325286865234
    },
    {
      "auxiliary_loss_clip": 0.01173633,
      "auxiliary_loss_mlp": 0.01032219,
      "balance_loss_clip": 0.97658122,
      "balance_loss_mlp": 1.02427673,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.4903868010492622,
      "language_loss": 0.80808365,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.83014214,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.708038091659546
    },
    {
      "auxiliary_loss_clip": 0.01185521,
      "auxiliary_loss_mlp": 0.01029935,
      "balance_loss_clip": 0.89929938,
      "balance_loss_mlp": 1.02156115,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 1.7278296760608853,
      "language_loss": 0.80106473,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.8232193,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 3.6717405319213867
    },
    {
      "auxiliary_loss_clip": 0.0118099,
      "auxiliary_loss_mlp": 0.0103419,
      "balance_loss_clip": 0.97386467,
      "balance_loss_mlp": 1.0264647,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.6727427629864602,
      "language_loss": 0.73610693,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.7582587,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.7594754695892334
    },
    {
      "auxiliary_loss_clip": 0.01166935,
      "auxiliary_loss_mlp": 0.01028886,
      "balance_loss_clip": 0.97514224,
      "balance_loss_mlp": 1.02157545,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 2.736630483390725,
      "language_loss": 0.76094216,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78290033,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.6470091342926025
    },
    {
      "auxiliary_loss_clip": 0.01174218,
      "auxiliary_loss_mlp": 0.01123436,
      "balance_loss_clip": 0.97035813,
      "balance_loss_mlp": 0.0,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.5936278619507065,
      "language_loss": 0.77904129,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.80201781,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.71950626373291
    },
    {
      "auxiliary_loss_clip": 0.01181453,
      "auxiliary_loss_mlp": 0.01025847,
      "balance_loss_clip": 1.05300319,
      "balance_loss_mlp": 1.01785362,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 1.8714139577679978,
      "language_loss": 0.78035223,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.80242527,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.6059887409210205
    },
    {
      "auxiliary_loss_clip": 0.01174191,
      "auxiliary_loss_mlp": 0.01030091,
      "balance_loss_clip": 0.97248936,
      "balance_loss_mlp": 1.02270555,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 1.4994643496197477,
      "language_loss": 0.82391608,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84595883,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.681539535522461
    },
    {
      "auxiliary_loss_clip": 0.01174093,
      "auxiliary_loss_mlp": 0.01026945,
      "balance_loss_clip": 1.01733804,
      "balance_loss_mlp": 1.02029967,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.590520664190833,
      "language_loss": 0.78786302,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.8098734,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.6236910820007324
    },
    {
      "auxiliary_loss_clip": 0.01183345,
      "auxiliary_loss_mlp": 0.0102889,
      "balance_loss_clip": 1.01325536,
      "balance_loss_mlp": 1.02143943,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 2.946087895218686,
      "language_loss": 0.84859371,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.87071598,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.7480289936065674
    },
    {
      "auxiliary_loss_clip": 0.01184676,
      "auxiliary_loss_mlp": 0.01027206,
      "balance_loss_clip": 1.0141207,
      "balance_loss_mlp": 1.01918888,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 2.162986611575255,
      "language_loss": 0.58927363,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61139244,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.8296260833740234
    },
    {
      "auxiliary_loss_clip": 0.0108814,
      "auxiliary_loss_mlp": 0.01000261,
      "balance_loss_clip": 0.95115596,
      "balance_loss_mlp": 0.99831784,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.8027607154715206,
      "language_loss": 0.60099459,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62187862,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.3551766872406006
    },
    {
      "auxiliary_loss_clip": 0.01175773,
      "auxiliary_loss_mlp": 0.01026292,
      "balance_loss_clip": 0.9731102,
      "balance_loss_mlp": 1.01815617,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 1.777114925943344,
      "language_loss": 0.76027095,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78229165,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 2.8067715167999268
    },
    {
      "auxiliary_loss_clip": 0.01182128,
      "auxiliary_loss_mlp": 0.01033594,
      "balance_loss_clip": 0.93776023,
      "balance_loss_mlp": 1.02623892,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 1.7249287264655002,
      "language_loss": 0.71657312,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.73873043,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.877586603164673
    },
    {
      "auxiliary_loss_clip": 0.01184876,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 0.93450773,
      "balance_loss_mlp": 1.01869535,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 1.659136541719576,
      "language_loss": 0.7065028,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.72862172,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 2.9336891174316406
    },
    {
      "auxiliary_loss_clip": 0.01085343,
      "auxiliary_loss_mlp": 0.01001204,
      "balance_loss_clip": 1.02570462,
      "balance_loss_mlp": 0.99927324,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6895151292941333,
      "language_loss": 0.63869095,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65955639,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 3.3457283973693848
    },
    {
      "auxiliary_loss_clip": 0.01177636,
      "auxiliary_loss_mlp": 0.01031345,
      "balance_loss_clip": 1.01595902,
      "balance_loss_mlp": 1.02410352,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.6108447533493309,
      "language_loss": 0.80732244,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.82941222,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 3.640289306640625
    },
    {
      "auxiliary_loss_clip": 0.01168373,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 0.93879485,
      "balance_loss_mlp": 1.02144277,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 1.8034165653291814,
      "language_loss": 0.61425757,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.63623214,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 2.684704065322876
    },
    {
      "auxiliary_loss_clip": 0.01173987,
      "auxiliary_loss_mlp": 0.0103605,
      "balance_loss_clip": 0.8994481,
      "balance_loss_mlp": 1.02725804,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 2.052183297157016,
      "language_loss": 0.81708521,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.8391856,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.675917387008667
    },
    {
      "auxiliary_loss_clip": 0.01177809,
      "auxiliary_loss_mlp": 0.01029345,
      "balance_loss_clip": 0.89987946,
      "balance_loss_mlp": 1.02129829,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 1.8245104425070762,
      "language_loss": 0.76967496,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.7917465,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 3.599838972091675
    },
    {
      "auxiliary_loss_clip": 0.01182984,
      "auxiliary_loss_mlp": 0.01022374,
      "balance_loss_clip": 1.01556373,
      "balance_loss_mlp": 1.01480389,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 1.4835468685335813,
      "language_loss": 0.77452266,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79657626,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 2.7072298526763916
    },
    {
      "auxiliary_loss_clip": 0.01083764,
      "auxiliary_loss_mlp": 0.01001145,
      "balance_loss_clip": 1.02441645,
      "balance_loss_mlp": 0.99915427,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8231680661438822,
      "language_loss": 0.53888184,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55973095,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 3.253556966781616
    },
    {
      "auxiliary_loss_clip": 0.01083612,
      "auxiliary_loss_mlp": 0.01002592,
      "balance_loss_clip": 1.02435064,
      "balance_loss_mlp": 1.00058961,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.857341537884657,
      "language_loss": 0.63326102,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65412307,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.2035772800445557
    },
    {
      "auxiliary_loss_clip": 0.01178943,
      "auxiliary_loss_mlp": 0.01029732,
      "balance_loss_clip": 1.01463509,
      "balance_loss_mlp": 1.02163732,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 2.409042995934821,
      "language_loss": 0.79913974,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.82122648,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.686555862426758
    },
    {
      "auxiliary_loss_clip": 0.01179369,
      "auxiliary_loss_mlp": 0.01028191,
      "balance_loss_clip": 0.97653866,
      "balance_loss_mlp": 1.02073479,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 1.603114029757432,
      "language_loss": 0.84259087,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86466646,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.707064151763916
    },
    {
      "auxiliary_loss_clip": 0.01178133,
      "auxiliary_loss_mlp": 0.01036279,
      "balance_loss_clip": 0.90315354,
      "balance_loss_mlp": 1.02807105,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.6197803460535811,
      "language_loss": 0.73035789,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75250196,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 3.7285141944885254
    },
    {
      "auxiliary_loss_clip": 0.01185597,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 1.01465821,
      "balance_loss_mlp": 1.01916325,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 1.9052707850249286,
      "language_loss": 0.78645551,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.80858976,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.679306745529175
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.01028138,
      "balance_loss_clip": 0.78345686,
      "balance_loss_mlp": 1.01969218,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 2.0054229204532077,
      "language_loss": 0.8395108,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.86148357,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 3.764838457107544
    },
    {
      "auxiliary_loss_clip": 0.01171683,
      "auxiliary_loss_mlp": 0.0102358,
      "balance_loss_clip": 0.93722844,
      "balance_loss_mlp": 1.01586151,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 1.8717355574804089,
      "language_loss": 0.82461154,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84656417,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.6967508792877197
    },
    {
      "auxiliary_loss_clip": 0.01180188,
      "auxiliary_loss_mlp": 0.01030592,
      "balance_loss_clip": 1.05259633,
      "balance_loss_mlp": 1.02279532,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.5889690120306155,
      "language_loss": 0.75460327,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77671105,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.657923936843872
    },
    {
      "auxiliary_loss_clip": 0.01182196,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.01696825,
      "balance_loss_mlp": 1.02367413,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.179337297228833,
      "language_loss": 0.79195422,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.81409562,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.6673030853271484
    },
    {
      "auxiliary_loss_clip": 0.01179894,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.05079842,
      "balance_loss_mlp": 1.02023983,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 1.4773446357934266,
      "language_loss": 0.8483063,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87038845,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.659348249435425
    },
    {
      "auxiliary_loss_clip": 0.01173213,
      "auxiliary_loss_mlp": 0.01028272,
      "balance_loss_clip": 0.93713903,
      "balance_loss_mlp": 1.02079773,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 2.0588243830436492,
      "language_loss": 0.80823225,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.8302471,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.7093605995178223
    },
    {
      "auxiliary_loss_clip": 0.01176224,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.01244164,
      "balance_loss_mlp": 1.01973403,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 1.7675224187009757,
      "language_loss": 0.72834283,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.75037515,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 2.6836512088775635
    },
    {
      "auxiliary_loss_clip": 0.01095361,
      "auxiliary_loss_mlp": 0.01002612,
      "balance_loss_clip": 0.96010602,
      "balance_loss_mlp": 1.0006808,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8360404100140526,
      "language_loss": 0.62440765,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64538729,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.2881152629852295
    },
    {
      "auxiliary_loss_clip": 0.01164403,
      "auxiliary_loss_mlp": 0.0103252,
      "balance_loss_clip": 0.78423238,
      "balance_loss_mlp": 1.0243721,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 2.0653932254706326,
      "language_loss": 0.68314403,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70511323,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 3.1311938762664795
    },
    {
      "auxiliary_loss_clip": 0.010827,
      "auxiliary_loss_mlp": 0.01002833,
      "balance_loss_clip": 1.0236938,
      "balance_loss_mlp": 1.00097311,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.8374719568843323,
      "language_loss": 0.60846627,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.62932158,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 3.3895885944366455
    },
    {
      "auxiliary_loss_clip": 0.01177969,
      "auxiliary_loss_mlp": 0.01026893,
      "balance_loss_clip": 0.93364954,
      "balance_loss_mlp": 1.01957917,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.098296424743388,
      "language_loss": 0.85448253,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.87653112,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.7412009239196777
    },
    {
      "auxiliary_loss_clip": 0.01178664,
      "auxiliary_loss_mlp": 0.01030619,
      "balance_loss_clip": 1.01502967,
      "balance_loss_mlp": 1.023067,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 1.6898086758072657,
      "language_loss": 0.78668922,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.8087821,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 2.7184863090515137
    },
    {
      "auxiliary_loss_clip": 0.01177301,
      "auxiliary_loss_mlp": 0.01029538,
      "balance_loss_clip": 1.01385522,
      "balance_loss_mlp": 1.02231944,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 1.7705394357459847,
      "language_loss": 0.76861084,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.79067928,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.6860299110412598
    },
    {
      "auxiliary_loss_clip": 0.01181287,
      "auxiliary_loss_mlp": 0.01122611,
      "balance_loss_clip": 1.01413131,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 1.6715264868848558,
      "language_loss": 0.78977191,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.8128109,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 2.6752169132232666
    },
    {
      "auxiliary_loss_clip": 0.01172089,
      "auxiliary_loss_mlp": 0.01032844,
      "balance_loss_clip": 0.97245711,
      "balance_loss_mlp": 1.02420759,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 2.374910899513716,
      "language_loss": 0.73421526,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75626457,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 3.049746513366699
    },
    {
      "auxiliary_loss_clip": 0.01179963,
      "auxiliary_loss_mlp": 0.01024309,
      "balance_loss_clip": 1.01475883,
      "balance_loss_mlp": 1.01701355,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 2.8360263220445856,
      "language_loss": 0.7138775,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.73592025,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 3.9099459648132324
    },
    {
      "auxiliary_loss_clip": 0.01163584,
      "auxiliary_loss_mlp": 0.01025646,
      "balance_loss_clip": 0.97099233,
      "balance_loss_mlp": 1.01793635,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 1.9423001157915287,
      "language_loss": 0.75616181,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.77805412,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 2.683242082595825
    },
    {
      "auxiliary_loss_clip": 0.01169499,
      "auxiliary_loss_mlp": 0.01024527,
      "balance_loss_clip": 0.9727962,
      "balance_loss_mlp": 1.01718915,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 1.761747108545789,
      "language_loss": 0.66134608,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68328637,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 2.715430974960327
    },
    {
      "auxiliary_loss_clip": 0.01159821,
      "auxiliary_loss_mlp": 0.01030417,
      "balance_loss_clip": 0.77882022,
      "balance_loss_mlp": 1.02283275,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.3125368560704926,
      "language_loss": 0.65507579,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.67697811,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 4.00312876701355
    },
    {
      "auxiliary_loss_clip": 0.01177886,
      "auxiliary_loss_mlp": 0.01027657,
      "balance_loss_clip": 1.01166165,
      "balance_loss_mlp": 1.02000415,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.5080421535909176,
      "language_loss": 0.75878388,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.78083932,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 2.867166042327881
    },
    {
      "auxiliary_loss_clip": 0.01176992,
      "auxiliary_loss_mlp": 0.01026001,
      "balance_loss_clip": 0.97621781,
      "balance_loss_mlp": 1.0188123,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 1.8205431688874485,
      "language_loss": 0.7913574,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81338727,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 2.7435343265533447
    },
    {
      "auxiliary_loss_clip": 0.01179607,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.05073988,
      "balance_loss_mlp": 1.01861811,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 1.8586306881911783,
      "language_loss": 0.66765082,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.68970954,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.6922383308410645
    },
    {
      "auxiliary_loss_clip": 0.01182519,
      "auxiliary_loss_mlp": 0.01032162,
      "balance_loss_clip": 1.01652288,
      "balance_loss_mlp": 1.02543592,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 1.8648414369170772,
      "language_loss": 0.68153703,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.70368385,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.6905221939086914
    },
    {
      "auxiliary_loss_clip": 0.01172043,
      "auxiliary_loss_mlp": 0.01034825,
      "balance_loss_clip": 0.97479582,
      "balance_loss_mlp": 1.02652764,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 1.8078607702813247,
      "language_loss": 0.68296421,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.70503289,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.7009339332580566
    },
    {
      "auxiliary_loss_clip": 0.01177003,
      "auxiliary_loss_mlp": 0.01026982,
      "balance_loss_clip": 1.01222849,
      "balance_loss_mlp": 1.01932311,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 2.3231938282104507,
      "language_loss": 0.67267108,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.69471091,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 3.9156887531280518
    },
    {
      "auxiliary_loss_clip": 0.01170734,
      "auxiliary_loss_mlp": 0.01022966,
      "balance_loss_clip": 0.97211099,
      "balance_loss_mlp": 1.01488936,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.1966714550577158,
      "language_loss": 0.65744889,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.6793859,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.8155155181884766
    },
    {
      "auxiliary_loss_clip": 0.0117968,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.05188155,
      "balance_loss_mlp": 1.01994181,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 1.9289562906507063,
      "language_loss": 0.75209606,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77416921,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 3.549952268600464
    },
    {
      "auxiliary_loss_clip": 0.01166716,
      "auxiliary_loss_mlp": 0.0103136,
      "balance_loss_clip": 0.89604318,
      "balance_loss_mlp": 1.0232414,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 1.9334940521132746,
      "language_loss": 0.77344185,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.79542267,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.7433714866638184
    },
    {
      "auxiliary_loss_clip": 0.01172786,
      "auxiliary_loss_mlp": 0.01036238,
      "balance_loss_clip": 0.97449839,
      "balance_loss_mlp": 1.0289005,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 1.9017644740098345,
      "language_loss": 0.83367163,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85576189,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.7334232330322266
    },
    {
      "auxiliary_loss_clip": 0.01180628,
      "auxiliary_loss_mlp": 0.01032525,
      "balance_loss_clip": 1.01410556,
      "balance_loss_mlp": 1.02429318,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 2.2851095292906334,
      "language_loss": 0.73605567,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.75818723,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.6750152111053467
    },
    {
      "auxiliary_loss_clip": 0.01180473,
      "auxiliary_loss_mlp": 0.01027654,
      "balance_loss_clip": 1.05304384,
      "balance_loss_mlp": 1.01947069,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 2.022984212717629,
      "language_loss": 0.73327315,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.7553544,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.630311965942383
    },
    {
      "auxiliary_loss_clip": 0.01171516,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 0.93698871,
      "balance_loss_mlp": 1.02134895,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.2378223865920344,
      "language_loss": 0.76168674,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78368783,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.720451593399048
    },
    {
      "auxiliary_loss_clip": 0.01178433,
      "auxiliary_loss_mlp": 0.01029818,
      "balance_loss_clip": 1.05209088,
      "balance_loss_mlp": 1.02255237,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 1.8694560186245304,
      "language_loss": 0.79955256,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82163501,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.5952701568603516
    },
    {
      "auxiliary_loss_clip": 0.01178264,
      "auxiliary_loss_mlp": 0.01025912,
      "balance_loss_clip": 1.05077457,
      "balance_loss_mlp": 1.01804423,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 1.741932616671515,
      "language_loss": 0.71695673,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.73899847,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.5855839252471924
    },
    {
      "auxiliary_loss_clip": 0.01178735,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.05234432,
      "balance_loss_mlp": 1.01879501,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 2.0696338484682224,
      "language_loss": 0.78513116,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80718136,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.6333892345428467
    },
    {
      "auxiliary_loss_clip": 0.01172995,
      "auxiliary_loss_mlp": 0.01028637,
      "balance_loss_clip": 0.97381413,
      "balance_loss_mlp": 1.0210073,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.474940125748584,
      "language_loss": 0.7800231,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.80203944,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.7074854373931885
    },
    {
      "auxiliary_loss_clip": 0.01179584,
      "auxiliary_loss_mlp": 0.01030598,
      "balance_loss_clip": 1.01449609,
      "balance_loss_mlp": 1.02231848,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 2.3726436118800383,
      "language_loss": 0.88793147,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.91003329,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.6841492652893066
    },
    {
      "auxiliary_loss_clip": 0.01185685,
      "auxiliary_loss_mlp": 0.01122915,
      "balance_loss_clip": 0.97612113,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 1.9176359553213722,
      "language_loss": 0.92569733,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.94878328,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 2.7545149326324463
    },
    {
      "auxiliary_loss_clip": 0.01171172,
      "auxiliary_loss_mlp": 0.01123065,
      "balance_loss_clip": 0.93862993,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 2.1248968261290697,
      "language_loss": 0.72905093,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.7519933,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.731682777404785
    },
    {
      "auxiliary_loss_clip": 0.01172709,
      "auxiliary_loss_mlp": 0.0102898,
      "balance_loss_clip": 0.97331351,
      "balance_loss_mlp": 1.02154732,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 1.8999935153328973,
      "language_loss": 0.76972675,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.79174358,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 2.6728670597076416
    },
    {
      "auxiliary_loss_clip": 0.01178593,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 0.90268248,
      "balance_loss_mlp": 1.02145994,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 1.576679524474797,
      "language_loss": 0.78073895,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80281675,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.895808219909668
    },
    {
      "auxiliary_loss_clip": 0.01183013,
      "auxiliary_loss_mlp": 0.01123134,
      "balance_loss_clip": 1.0141753,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 2.049127204466992,
      "language_loss": 0.82974696,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.85280842,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 3.813405752182007
    },
    {
      "auxiliary_loss_clip": 0.01177269,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 0.97592896,
      "balance_loss_mlp": 1.020908,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 2.1569500690350085,
      "language_loss": 0.88711727,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.90916908,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 2.7839815616607666
    },
    {
      "auxiliary_loss_clip": 0.01177857,
      "auxiliary_loss_mlp": 0.01027115,
      "balance_loss_clip": 1.01449239,
      "balance_loss_mlp": 1.01897264,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 2.3677188487423804,
      "language_loss": 0.76409596,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78614569,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.6011080741882324
    },
    {
      "auxiliary_loss_clip": 0.01086139,
      "auxiliary_loss_mlp": 0.01000395,
      "balance_loss_clip": 0.95172143,
      "balance_loss_mlp": 0.99845171,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8368251101518631,
      "language_loss": 0.6076355,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62850082,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 4.076733589172363
    },
    {
      "auxiliary_loss_clip": 0.01182881,
      "auxiliary_loss_mlp": 0.01023484,
      "balance_loss_clip": 0.93705302,
      "balance_loss_mlp": 1.01627517,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 1.8098982164853743,
      "language_loss": 0.7416901,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76375371,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.6503756046295166
    },
    {
      "auxiliary_loss_clip": 0.01180024,
      "auxiliary_loss_mlp": 0.01028257,
      "balance_loss_clip": 1.01603878,
      "balance_loss_mlp": 1.02074349,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.0790670382211185,
      "language_loss": 0.64219826,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66428107,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 2.60996413230896
    },
    {
      "auxiliary_loss_clip": 0.01183082,
      "auxiliary_loss_mlp": 0.0102243,
      "balance_loss_clip": 1.01584196,
      "balance_loss_mlp": 1.01473486,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 1.7971696628897893,
      "language_loss": 0.76468855,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.7867437,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 2.6286261081695557
    },
    {
      "auxiliary_loss_clip": 0.01082588,
      "auxiliary_loss_mlp": 0.00999784,
      "balance_loss_clip": 0.98792613,
      "balance_loss_mlp": 0.99792403,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8339636326111658,
      "language_loss": 0.57637185,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59719557,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.202253818511963
    },
    {
      "auxiliary_loss_clip": 0.01178555,
      "auxiliary_loss_mlp": 0.01122426,
      "balance_loss_clip": 0.97924042,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 2.2977027989890213,
      "language_loss": 0.80088323,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.82389307,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.6478488445281982
    },
    {
      "auxiliary_loss_clip": 0.0118338,
      "auxiliary_loss_mlp": 0.01031603,
      "balance_loss_clip": 0.97335267,
      "balance_loss_mlp": 1.0239799,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 2.5958403079984036,
      "language_loss": 0.70735663,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.72950649,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 3.5354270935058594
    },
    {
      "auxiliary_loss_clip": 0.01176776,
      "auxiliary_loss_mlp": 0.01033227,
      "balance_loss_clip": 1.01329267,
      "balance_loss_mlp": 1.02597308,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 2.323039433326054,
      "language_loss": 0.83985078,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.86195076,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.64373517036438
    },
    {
      "auxiliary_loss_clip": 0.01174217,
      "auxiliary_loss_mlp": 0.01025761,
      "balance_loss_clip": 1.01342165,
      "balance_loss_mlp": 1.01826859,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 1.8704216059035683,
      "language_loss": 0.78563392,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.8076337,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 3.5923702716827393
    },
    {
      "auxiliary_loss_clip": 0.01181443,
      "auxiliary_loss_mlp": 0.01031664,
      "balance_loss_clip": 1.01339912,
      "balance_loss_mlp": 1.02409768,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 1.6680309506424476,
      "language_loss": 0.8633216,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88545263,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.7869973182678223
    },
    {
      "auxiliary_loss_clip": 0.01171096,
      "auxiliary_loss_mlp": 0.01030237,
      "balance_loss_clip": 0.97443837,
      "balance_loss_mlp": 1.0221132,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 1.915399665637628,
      "language_loss": 0.66341102,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68542433,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.7864463329315186
    },
    {
      "auxiliary_loss_clip": 0.01088742,
      "auxiliary_loss_mlp": 0.01003705,
      "balance_loss_clip": 0.9888767,
      "balance_loss_mlp": 1.00180936,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.8445768424505303,
      "language_loss": 0.59568948,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61661398,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.1675055027008057
    },
    {
      "auxiliary_loss_clip": 0.01178472,
      "auxiliary_loss_mlp": 0.01027643,
      "balance_loss_clip": 1.01471782,
      "balance_loss_mlp": 1.01981056,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 2.0173410510006,
      "language_loss": 0.76813245,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.79019362,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.721036911010742
    },
    {
      "auxiliary_loss_clip": 0.01178902,
      "auxiliary_loss_mlp": 0.01024267,
      "balance_loss_clip": 1.01333761,
      "balance_loss_mlp": 1.01670957,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 1.8055564090813778,
      "language_loss": 0.83015895,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85219073,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.661827802658081
    },
    {
      "auxiliary_loss_clip": 0.01180234,
      "auxiliary_loss_mlp": 0.01031526,
      "balance_loss_clip": 1.0516355,
      "balance_loss_mlp": 1.02460623,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 2.0040788347323004,
      "language_loss": 0.80833769,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.8304553,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.6305954456329346
    },
    {
      "auxiliary_loss_clip": 0.01170922,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.01258481,
      "balance_loss_mlp": 1.01850569,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 1.7659028303097246,
      "language_loss": 0.67149824,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.69347125,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.6175570487976074
    },
    {
      "auxiliary_loss_clip": 0.0118569,
      "auxiliary_loss_mlp": 0.01025502,
      "balance_loss_clip": 1.01313114,
      "balance_loss_mlp": 1.01724064,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 2.5482121544366723,
      "language_loss": 0.75751704,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77962899,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.7296864986419678
    },
    {
      "auxiliary_loss_clip": 0.01166513,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.01000059,
      "balance_loss_mlp": 1.0187242,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 1.5317066585117391,
      "language_loss": 0.62243283,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.64435393,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.7079062461853027
    },
    {
      "auxiliary_loss_clip": 0.01169266,
      "auxiliary_loss_mlp": 0.01029684,
      "balance_loss_clip": 0.89944434,
      "balance_loss_mlp": 1.02176905,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 2.274220176960059,
      "language_loss": 0.84018338,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.8621729,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.730999231338501
    },
    {
      "auxiliary_loss_clip": 0.01177278,
      "auxiliary_loss_mlp": 0.01025622,
      "balance_loss_clip": 1.01404667,
      "balance_loss_mlp": 1.01845169,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 1.8628885528550252,
      "language_loss": 0.78758144,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.80961049,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 2.6440250873565674
    },
    {
      "auxiliary_loss_clip": 0.01173654,
      "auxiliary_loss_mlp": 0.01024487,
      "balance_loss_clip": 1.01229,
      "balance_loss_mlp": 1.01717925,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 1.8552068204636436,
      "language_loss": 0.81089169,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83287311,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.6920578479766846
    },
    {
      "auxiliary_loss_clip": 0.01173808,
      "auxiliary_loss_mlp": 0.010267,
      "balance_loss_clip": 1.0103997,
      "balance_loss_mlp": 1.01882625,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 5.269708855888101,
      "language_loss": 0.75704652,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.7790516,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 2.635894775390625
    },
    {
      "auxiliary_loss_clip": 0.01178801,
      "auxiliary_loss_mlp": 0.01028745,
      "balance_loss_clip": 0.93808031,
      "balance_loss_mlp": 1.02145243,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 1.8043793025899886,
      "language_loss": 0.7615605,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78363597,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.729919910430908
    },
    {
      "auxiliary_loss_clip": 0.01171026,
      "auxiliary_loss_mlp": 0.0102855,
      "balance_loss_clip": 0.85788864,
      "balance_loss_mlp": 1.02036595,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 2.081512821353048,
      "language_loss": 0.76808429,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.79008007,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 3.93312406539917
    },
    {
      "auxiliary_loss_clip": 0.01176326,
      "auxiliary_loss_mlp": 0.01026762,
      "balance_loss_clip": 0.97554231,
      "balance_loss_mlp": 1.01844096,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 1.4737605152804958,
      "language_loss": 0.81370008,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83573091,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.929112434387207
    },
    {
      "auxiliary_loss_clip": 0.01082715,
      "auxiliary_loss_mlp": 0.01003048,
      "balance_loss_clip": 0.95048159,
      "balance_loss_mlp": 1.00117671,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8281621713429569,
      "language_loss": 0.5826053,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.60346287,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.2506000995635986
    },
    {
      "auxiliary_loss_clip": 0.01174606,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 0.93441665,
      "balance_loss_mlp": 1.01783895,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.3513164551559766,
      "language_loss": 0.7539714,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.77597249,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 3.7685980796813965
    },
    {
      "auxiliary_loss_clip": 0.01170916,
      "auxiliary_loss_mlp": 0.01028162,
      "balance_loss_clip": 0.93361646,
      "balance_loss_mlp": 1.02053261,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.5768730397576103,
      "language_loss": 0.77216959,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79416037,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.7548093795776367
    },
    {
      "auxiliary_loss_clip": 0.0117083,
      "auxiliary_loss_mlp": 0.0103033,
      "balance_loss_clip": 1.01046944,
      "balance_loss_mlp": 1.02262354,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 2.4420038439330916,
      "language_loss": 0.78696465,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.80897623,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 2.654402494430542
    },
    {
      "auxiliary_loss_clip": 0.01165535,
      "auxiliary_loss_mlp": 0.01022459,
      "balance_loss_clip": 0.97059703,
      "balance_loss_mlp": 1.01525021,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 1.5426630677368576,
      "language_loss": 0.79935175,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82123172,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 2.654046058654785
    },
    {
      "auxiliary_loss_clip": 0.0117535,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 0.93554533,
      "balance_loss_mlp": 1.02009821,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 1.8647457459779042,
      "language_loss": 0.73547131,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.757496,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 2.7211925983428955
    },
    {
      "auxiliary_loss_clip": 0.01184054,
      "auxiliary_loss_mlp": 0.01034225,
      "balance_loss_clip": 0.97756302,
      "balance_loss_mlp": 1.02620268,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 1.9807819943303764,
      "language_loss": 0.7163533,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73853612,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 3.6406452655792236
    },
    {
      "auxiliary_loss_clip": 0.01179546,
      "auxiliary_loss_mlp": 0.01026446,
      "balance_loss_clip": 1.01144087,
      "balance_loss_mlp": 1.01892662,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 2.5078438563994525,
      "language_loss": 0.69583064,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.71789056,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.5896196365356445
    },
    {
      "auxiliary_loss_clip": 0.01167828,
      "auxiliary_loss_mlp": 0.01032804,
      "balance_loss_clip": 1.01169229,
      "balance_loss_mlp": 1.02560973,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 2.8185280347683226,
      "language_loss": 0.76487267,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.786879,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 3.5796875953674316
    },
    {
      "auxiliary_loss_clip": 0.01180237,
      "auxiliary_loss_mlp": 0.01032719,
      "balance_loss_clip": 0.97740608,
      "balance_loss_mlp": 1.02541769,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 1.7626853923622143,
      "language_loss": 0.80092341,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.823053,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.621946096420288
    },
    {
      "auxiliary_loss_clip": 0.01169223,
      "auxiliary_loss_mlp": 0.01028648,
      "balance_loss_clip": 0.89459842,
      "balance_loss_mlp": 1.02063143,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 1.7950195195035714,
      "language_loss": 0.82854915,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.85052776,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.7966928482055664
    },
    {
      "auxiliary_loss_clip": 0.01172486,
      "auxiliary_loss_mlp": 0.01029107,
      "balance_loss_clip": 0.97158223,
      "balance_loss_mlp": 1.0209825,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 1.825879298399853,
      "language_loss": 0.71106422,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.73308015,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.762299060821533
    },
    {
      "auxiliary_loss_clip": 0.01166147,
      "auxiliary_loss_mlp": 0.01027361,
      "balance_loss_clip": 0.9743073,
      "balance_loss_mlp": 1.02037823,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 1.8418302815786745,
      "language_loss": 0.83677322,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.85870826,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.6896393299102783
    },
    {
      "auxiliary_loss_clip": 0.01177268,
      "auxiliary_loss_mlp": 0.0102849,
      "balance_loss_clip": 0.97446805,
      "balance_loss_mlp": 1.02079177,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 1.8932330464808924,
      "language_loss": 0.84787202,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.86992967,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.687657117843628
    },
    {
      "auxiliary_loss_clip": 0.01173793,
      "auxiliary_loss_mlp": 0.01031772,
      "balance_loss_clip": 0.97575492,
      "balance_loss_mlp": 1.02461362,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 1.8265315138822675,
      "language_loss": 0.80418515,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82624084,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.695629835128784
    },
    {
      "auxiliary_loss_clip": 0.01179102,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.05183005,
      "balance_loss_mlp": 1.0193764,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.5552975844561523,
      "language_loss": 0.62776554,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64982826,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.677269697189331
    },
    {
      "auxiliary_loss_clip": 0.0117785,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 0.97720081,
      "balance_loss_mlp": 1.02066374,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 3.036822237607028,
      "language_loss": 0.69270647,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.71476912,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.7271313667297363
    },
    {
      "auxiliary_loss_clip": 0.01173057,
      "auxiliary_loss_mlp": 0.0102796,
      "balance_loss_clip": 0.85792196,
      "balance_loss_mlp": 1.01940084,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 2.225124051974836,
      "language_loss": 0.73787111,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.75988132,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.8412020206451416
    },
    {
      "auxiliary_loss_clip": 0.01182932,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 0.97348106,
      "balance_loss_mlp": 1.02273142,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 6.549200929290353,
      "language_loss": 0.78373486,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80587453,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.8347532749176025
    },
    {
      "auxiliary_loss_clip": 0.0108362,
      "auxiliary_loss_mlp": 0.01007287,
      "balance_loss_clip": 0.98523021,
      "balance_loss_mlp": 1.00546288,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7168382879251968,
      "language_loss": 0.58994943,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.6108585,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 3.342674732208252
    },
    {
      "auxiliary_loss_clip": 0.01175107,
      "auxiliary_loss_mlp": 0.01025495,
      "balance_loss_clip": 1.01422811,
      "balance_loss_mlp": 1.01778173,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 1.6129533903907167,
      "language_loss": 0.74516106,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76716715,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 2.712252140045166
    },
    {
      "auxiliary_loss_clip": 0.01185025,
      "auxiliary_loss_mlp": 0.01027387,
      "balance_loss_clip": 1.01623893,
      "balance_loss_mlp": 1.01965666,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.123438624107982,
      "language_loss": 0.74336815,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76549226,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 2.6301887035369873
    },
    {
      "auxiliary_loss_clip": 0.01085942,
      "auxiliary_loss_mlp": 0.01116403,
      "balance_loss_clip": 0.91267687,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7642682826861105,
      "language_loss": 0.58003616,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.6020596,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 3.2508060932159424
    },
    {
      "auxiliary_loss_clip": 0.01178648,
      "auxiliary_loss_mlp": 0.0102722,
      "balance_loss_clip": 1.01360774,
      "balance_loss_mlp": 1.01926899,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 1.742924871037084,
      "language_loss": 0.78548455,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.80754322,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.5660741329193115
    },
    {
      "auxiliary_loss_clip": 0.01170138,
      "auxiliary_loss_mlp": 0.01027371,
      "balance_loss_clip": 0.97101426,
      "balance_loss_mlp": 1.0195632,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.4842416972279542,
      "language_loss": 0.79598016,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.81795526,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 3.6996726989746094
    },
    {
      "auxiliary_loss_clip": 0.01186398,
      "auxiliary_loss_mlp": 0.0102564,
      "balance_loss_clip": 1.0151279,
      "balance_loss_mlp": 1.0178318,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 1.8131908238161927,
      "language_loss": 0.78741401,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.80953443,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.6979641914367676
    },
    {
      "auxiliary_loss_clip": 0.01083497,
      "auxiliary_loss_mlp": 0.01004624,
      "balance_loss_clip": 0.90866041,
      "balance_loss_mlp": 1.00274014,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 1.0092126819043963,
      "language_loss": 0.59999585,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.62087703,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 3.3164443969726562
    },
    {
      "auxiliary_loss_clip": 0.01180178,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.05261445,
      "balance_loss_mlp": 1.01590395,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 2.3892092638927256,
      "language_loss": 0.7512219,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.77326065,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 3.658742666244507
    },
    {
      "auxiliary_loss_clip": 0.01173604,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 0.97266752,
      "balance_loss_mlp": 1.02190888,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 1.9042885127774343,
      "language_loss": 0.81961513,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.84164274,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.684547185897827
    },
    {
      "auxiliary_loss_clip": 0.01178579,
      "auxiliary_loss_mlp": 0.01025461,
      "balance_loss_clip": 1.01258314,
      "balance_loss_mlp": 1.01749802,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 1.6747749343251748,
      "language_loss": 0.81451154,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83655202,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.649697780609131
    },
    {
      "auxiliary_loss_clip": 0.01182898,
      "auxiliary_loss_mlp": 0.01030628,
      "balance_loss_clip": 0.93658966,
      "balance_loss_mlp": 1.02255189,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 1.726425132415522,
      "language_loss": 0.71831226,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.74044752,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 2.7100472450256348
    },
    {
      "auxiliary_loss_clip": 0.01178306,
      "auxiliary_loss_mlp": 0.01032334,
      "balance_loss_clip": 1.05023885,
      "balance_loss_mlp": 1.02457619,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.5862006545456424,
      "language_loss": 0.73775983,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.75986618,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 2.552375555038452
    },
    {
      "auxiliary_loss_clip": 0.01170002,
      "auxiliary_loss_mlp": 0.01025927,
      "balance_loss_clip": 0.97193933,
      "balance_loss_mlp": 1.01796341,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 1.7626861393265931,
      "language_loss": 0.72490537,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.74686468,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 3.5390467643737793
    },
    {
      "auxiliary_loss_clip": 0.01182274,
      "auxiliary_loss_mlp": 0.01027799,
      "balance_loss_clip": 0.97514653,
      "balance_loss_mlp": 1.02010727,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 2.1140228330961266,
      "language_loss": 0.79945409,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.82155484,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.6528232097625732
    },
    {
      "auxiliary_loss_clip": 0.01177189,
      "auxiliary_loss_mlp": 0.01027141,
      "balance_loss_clip": 0.97419244,
      "balance_loss_mlp": 1.01935661,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 1.688842309737281,
      "language_loss": 0.67779171,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.69983506,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 3.7060768604278564
    },
    {
      "auxiliary_loss_clip": 0.01165748,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 0.97449285,
      "balance_loss_mlp": 1.01861286,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 2.7528317202091306,
      "language_loss": 0.6875596,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.70948195,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.768073320388794
    },
    {
      "auxiliary_loss_clip": 0.01175578,
      "auxiliary_loss_mlp": 0.01027434,
      "balance_loss_clip": 0.93694711,
      "balance_loss_mlp": 1.0190891,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 1.7317922988567316,
      "language_loss": 0.76441544,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.78644556,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.684333086013794
    },
    {
      "auxiliary_loss_clip": 0.01175419,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 0.93791473,
      "balance_loss_mlp": 1.02216709,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 2.179863196788815,
      "language_loss": 0.80715382,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.82921374,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.760343313217163
    },
    {
      "auxiliary_loss_clip": 0.01175235,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 0.89977568,
      "balance_loss_mlp": 1.02327681,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.6247674571985953,
      "language_loss": 0.78977823,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81184173,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.811763048171997
    },
    {
      "auxiliary_loss_clip": 0.01103354,
      "auxiliary_loss_mlp": 0.01003548,
      "balance_loss_clip": 0.88687116,
      "balance_loss_mlp": 1.00172448,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.7743373287857619,
      "language_loss": 0.5483886,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56945765,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.3414509296417236
    },
    {
      "auxiliary_loss_clip": 0.01180126,
      "auxiliary_loss_mlp": 0.01024542,
      "balance_loss_clip": 1.01253438,
      "balance_loss_mlp": 1.01672816,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 1.8066097013422124,
      "language_loss": 0.78790832,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.809955,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.634348154067993
    },
    {
      "auxiliary_loss_clip": 0.01170923,
      "auxiliary_loss_mlp": 0.0102751,
      "balance_loss_clip": 0.85263389,
      "balance_loss_mlp": 1.01937962,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 1.9012827104769112,
      "language_loss": 0.63863307,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66061735,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 2.800853729248047
    },
    {
      "auxiliary_loss_clip": 0.01168416,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 0.97464263,
      "balance_loss_mlp": 1.01925921,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 2.184572312175384,
      "language_loss": 0.78057748,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80252486,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.6830201148986816
    },
    {
      "auxiliary_loss_clip": 0.01186007,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 1.0566082,
      "balance_loss_mlp": 1.02427828,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 2.272007949250053,
      "language_loss": 0.75362831,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.77580798,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.5979490280151367
    },
    {
      "auxiliary_loss_clip": 0.01186083,
      "auxiliary_loss_mlp": 0.01027686,
      "balance_loss_clip": 1.01681721,
      "balance_loss_mlp": 1.01960993,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 2.207917266449237,
      "language_loss": 0.82600129,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84813893,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.6260414123535156
    },
    {
      "auxiliary_loss_clip": 0.01173821,
      "auxiliary_loss_mlp": 0.01028733,
      "balance_loss_clip": 0.97314262,
      "balance_loss_mlp": 1.02118731,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 1.6457028969998406,
      "language_loss": 0.64063716,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.66266268,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.6899003982543945
    },
    {
      "auxiliary_loss_clip": 0.01175204,
      "auxiliary_loss_mlp": 0.01030263,
      "balance_loss_clip": 0.93880475,
      "balance_loss_mlp": 1.02251399,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 3.67224633657248,
      "language_loss": 0.78294075,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80499542,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 2.80299973487854
    },
    {
      "auxiliary_loss_clip": 0.01173742,
      "auxiliary_loss_mlp": 0.01030677,
      "balance_loss_clip": 0.97815943,
      "balance_loss_mlp": 1.02320838,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 2.228999410625891,
      "language_loss": 0.87521654,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.89726073,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 2.662658214569092
    },
    {
      "auxiliary_loss_clip": 0.01084458,
      "auxiliary_loss_mlp": 0.0099975,
      "balance_loss_clip": 0.98673087,
      "balance_loss_mlp": 0.99809277,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.8651603449748272,
      "language_loss": 0.55971205,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.58055413,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 3.227248191833496
    },
    {
      "auxiliary_loss_clip": 0.01169794,
      "auxiliary_loss_mlp": 0.01026206,
      "balance_loss_clip": 0.97293717,
      "balance_loss_mlp": 1.01867151,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 2.2018236728839877,
      "language_loss": 0.72398162,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74594164,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.6913743019104004
    },
    {
      "auxiliary_loss_clip": 0.01174063,
      "auxiliary_loss_mlp": 0.01026308,
      "balance_loss_clip": 1.01334643,
      "balance_loss_mlp": 1.01889944,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 1.6273348067854687,
      "language_loss": 0.75168282,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77368653,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 3.671506643295288
    },
    {
      "auxiliary_loss_clip": 0.01178709,
      "auxiliary_loss_mlp": 0.01027238,
      "balance_loss_clip": 0.97738147,
      "balance_loss_mlp": 1.01995468,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 1.9174899422228897,
      "language_loss": 0.70648229,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.72854173,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.683335304260254
    },
    {
      "auxiliary_loss_clip": 0.01165996,
      "auxiliary_loss_mlp": 0.01027297,
      "balance_loss_clip": 0.97265685,
      "balance_loss_mlp": 1.01995337,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 1.697058197190965,
      "language_loss": 0.77130777,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.79324067,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.72682785987854
    },
    {
      "auxiliary_loss_clip": 0.01176787,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 1.0148741,
      "balance_loss_mlp": 1.01821947,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 1.9421942923411086,
      "language_loss": 0.69840968,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.72043002,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 3.684110641479492
    },
    {
      "auxiliary_loss_clip": 0.01166343,
      "auxiliary_loss_mlp": 0.01031477,
      "balance_loss_clip": 0.97165382,
      "balance_loss_mlp": 1.02422285,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 1.7862306386939601,
      "language_loss": 0.90436196,
      "learning_rate": 2e-06,
      "loss": 0.92634022,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.7229526042938232
    },
    {
      "auxiliary_loss_clip": 0.01178718,
      "auxiliary_loss_mlp": 0.01029636,
      "balance_loss_clip": 1.0538702,
      "balance_loss_mlp": 1.02217305,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 1.6610035961897442,
      "language_loss": 0.85774225,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.87982577,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.636569023132324
    },
    {
      "auxiliary_loss_clip": 0.01178301,
      "auxiliary_loss_mlp": 0.01031817,
      "balance_loss_clip": 1.05365705,
      "balance_loss_mlp": 1.02470016,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.445576059830181,
      "language_loss": 0.7916826,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81378382,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.6551122665405273
    },
    {
      "auxiliary_loss_clip": 0.01179781,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.01461792,
      "balance_loss_mlp": 1.02153909,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 1.8488870237807173,
      "language_loss": 0.78474957,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.80683655,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 2.601257562637329
    },
    {
      "auxiliary_loss_clip": 0.01168785,
      "auxiliary_loss_mlp": 0.01022997,
      "balance_loss_clip": 0.97039264,
      "balance_loss_mlp": 1.01597595,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 1.9384856461165818,
      "language_loss": 0.74248642,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.7644043,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 3.6052112579345703
    },
    {
      "auxiliary_loss_clip": 0.01169793,
      "auxiliary_loss_mlp": 0.01122991,
      "balance_loss_clip": 0.85937643,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.535040050006439,
      "language_loss": 0.71395516,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.73688304,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 2.7742114067077637
    },
    {
      "auxiliary_loss_clip": 0.01178202,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.01512802,
      "balance_loss_mlp": 1.01867676,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 1.8319600203841768,
      "language_loss": 0.78579277,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.80783594,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 3.5202107429504395
    },
    {
      "auxiliary_loss_clip": 0.01174917,
      "auxiliary_loss_mlp": 0.01122122,
      "balance_loss_clip": 1.0118556,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.9803702270840382,
      "language_loss": 0.72819269,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.75116313,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.6698014736175537
    },
    {
      "auxiliary_loss_clip": 0.01173679,
      "auxiliary_loss_mlp": 0.01024816,
      "balance_loss_clip": 0.97524524,
      "balance_loss_mlp": 1.01721668,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 1.9802525086978704,
      "language_loss": 0.6752255,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69721043,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.6725354194641113
    },
    {
      "auxiliary_loss_clip": 0.01170656,
      "auxiliary_loss_mlp": 0.01028947,
      "balance_loss_clip": 0.97296959,
      "balance_loss_mlp": 1.02149022,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 2.555539599623418,
      "language_loss": 0.67067623,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.69267231,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.7176592350006104
    },
    {
      "auxiliary_loss_clip": 0.01172213,
      "auxiliary_loss_mlp": 0.01026406,
      "balance_loss_clip": 0.97189796,
      "balance_loss_mlp": 1.01876426,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 2.6984228881415673,
      "language_loss": 0.68003476,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.702021,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.6122829914093018
    },
    {
      "auxiliary_loss_clip": 0.01178886,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 0.97399884,
      "balance_loss_mlp": 1.02224886,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 1.4897291739503407,
      "language_loss": 0.8089906,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.83107382,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.763117790222168
    },
    {
      "auxiliary_loss_clip": 0.01175621,
      "auxiliary_loss_mlp": 0.01023368,
      "balance_loss_clip": 0.9762398,
      "balance_loss_mlp": 1.01626635,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 1.6535070299152925,
      "language_loss": 0.79044789,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.81243777,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.743835210800171
    },
    {
      "auxiliary_loss_clip": 0.01179703,
      "auxiliary_loss_mlp": 0.01029656,
      "balance_loss_clip": 1.05120063,
      "balance_loss_mlp": 1.02130508,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 2.2995513560834997,
      "language_loss": 0.7732209,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.79531455,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 2.729722499847412
    },
    {
      "auxiliary_loss_clip": 0.01173439,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 0.82013178,
      "balance_loss_mlp": 1.01805663,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.4444002969499907,
      "language_loss": 0.80289131,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.82488424,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.932920217514038
    },
    {
      "auxiliary_loss_clip": 0.01173491,
      "auxiliary_loss_mlp": 0.01028485,
      "balance_loss_clip": 1.01293957,
      "balance_loss_mlp": 1.0212338,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 1.9109804843503875,
      "language_loss": 0.76519096,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.7872107,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 2.7472758293151855
    },
    {
      "auxiliary_loss_clip": 0.01174939,
      "auxiliary_loss_mlp": 0.01023906,
      "balance_loss_clip": 1.04925799,
      "balance_loss_mlp": 1.01627648,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 4.353622204573165,
      "language_loss": 0.83347309,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85546148,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.75736403465271
    },
    {
      "auxiliary_loss_clip": 0.01176175,
      "auxiliary_loss_mlp": 0.01122435,
      "balance_loss_clip": 1.01441908,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 2.1331192355035378,
      "language_loss": 0.84770143,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.8706876,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.647461414337158
    },
    {
      "auxiliary_loss_clip": 0.01179988,
      "auxiliary_loss_mlp": 0.01034727,
      "balance_loss_clip": 1.01427197,
      "balance_loss_mlp": 1.02701163,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 2.70834214414083,
      "language_loss": 0.85720068,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.8793478,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 2.6905131340026855
    },
    {
      "auxiliary_loss_clip": 0.01177313,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 0.93637347,
      "balance_loss_mlp": 1.0229702,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 3.290877381282161,
      "language_loss": 0.66673172,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.68881279,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 2.723557710647583
    },
    {
      "auxiliary_loss_clip": 0.01183262,
      "auxiliary_loss_mlp": 0.01034758,
      "balance_loss_clip": 1.01480794,
      "balance_loss_mlp": 1.02721238,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.0278714867106222,
      "language_loss": 0.7438699,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.7660501,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.604175329208374
    },
    {
      "auxiliary_loss_clip": 0.01172445,
      "auxiliary_loss_mlp": 0.01030983,
      "balance_loss_clip": 0.97394109,
      "balance_loss_mlp": 1.02371693,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 1.653951880844074,
      "language_loss": 0.68622768,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.70826197,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 2.814754009246826
    },
    {
      "auxiliary_loss_clip": 0.01090016,
      "auxiliary_loss_mlp": 0.00999815,
      "balance_loss_clip": 0.95070779,
      "balance_loss_mlp": 0.99801457,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.7484159707873345,
      "language_loss": 0.58757734,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.60847563,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 4.397078514099121
    },
    {
      "auxiliary_loss_clip": 0.01091933,
      "auxiliary_loss_mlp": 0.01001577,
      "balance_loss_clip": 0.95117408,
      "balance_loss_mlp": 0.99969333,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.769065538021447,
      "language_loss": 0.54725111,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56818616,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.2814226150512695
    },
    {
      "auxiliary_loss_clip": 0.01171589,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 0.9768306,
      "balance_loss_mlp": 1.02096415,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.5666143895383697,
      "language_loss": 0.82597184,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.84796941,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.758751392364502
    },
    {
      "auxiliary_loss_clip": 0.01168189,
      "auxiliary_loss_mlp": 0.01034653,
      "balance_loss_clip": 0.9321906,
      "balance_loss_mlp": 1.02709532,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.0736288974757437,
      "language_loss": 0.8195858,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.84161425,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 3.6291232109069824
    },
    {
      "auxiliary_loss_clip": 0.01177246,
      "auxiliary_loss_mlp": 0.0102787,
      "balance_loss_clip": 1.0140121,
      "balance_loss_mlp": 1.02056563,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 1.9558940119797772,
      "language_loss": 0.86238205,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88443315,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.6591126918792725
    },
    {
      "auxiliary_loss_clip": 0.01178595,
      "auxiliary_loss_mlp": 0.0103025,
      "balance_loss_clip": 1.01629508,
      "balance_loss_mlp": 1.02219748,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 3.9317112575735242,
      "language_loss": 0.77405655,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.79614496,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.747605800628662
    },
    {
      "auxiliary_loss_clip": 0.0117017,
      "auxiliary_loss_mlp": 0.01031653,
      "balance_loss_clip": 0.89828956,
      "balance_loss_mlp": 1.02431524,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 1.48962459162743,
      "language_loss": 0.72425586,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74627405,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.7882142066955566
    },
    {
      "auxiliary_loss_clip": 0.01178931,
      "auxiliary_loss_mlp": 0.01029118,
      "balance_loss_clip": 1.05298448,
      "balance_loss_mlp": 1.02159607,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 2.19900396706331,
      "language_loss": 0.76929373,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.79137421,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.6668152809143066
    },
    {
      "auxiliary_loss_clip": 0.01178333,
      "auxiliary_loss_mlp": 0.01028606,
      "balance_loss_clip": 0.90363872,
      "balance_loss_mlp": 1.02017176,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.4648744228335915,
      "language_loss": 0.76308441,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78515381,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 3.682837724685669
    },
    {
      "auxiliary_loss_clip": 0.01163656,
      "auxiliary_loss_mlp": 0.01027673,
      "balance_loss_clip": 0.89436734,
      "balance_loss_mlp": 1.02025819,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 3.629076597650621,
      "language_loss": 0.6962921,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71820545,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 2.6838808059692383
    },
    {
      "auxiliary_loss_clip": 0.01178774,
      "auxiliary_loss_mlp": 0.01026223,
      "balance_loss_clip": 1.01472199,
      "balance_loss_mlp": 1.01782489,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 2.311756960680654,
      "language_loss": 0.70590925,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72795916,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 3.5766732692718506
    },
    {
      "auxiliary_loss_clip": 0.01179019,
      "auxiliary_loss_mlp": 0.01028102,
      "balance_loss_clip": 0.97780573,
      "balance_loss_mlp": 1.02063906,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 2.3403136391261072,
      "language_loss": 0.79057944,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.81265068,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.6998822689056396
    },
    {
      "auxiliary_loss_clip": 0.01181803,
      "auxiliary_loss_mlp": 0.01031798,
      "balance_loss_clip": 1.01372361,
      "balance_loss_mlp": 1.02378154,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 1.7480174116692258,
      "language_loss": 0.67125463,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69339061,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.6633734703063965
    },
    {
      "auxiliary_loss_clip": 0.01179383,
      "auxiliary_loss_mlp": 0.01030257,
      "balance_loss_clip": 0.97412264,
      "balance_loss_mlp": 1.02251124,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 1.7177461205122788,
      "language_loss": 0.841268,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.8633644,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.7499139308929443
    },
    {
      "auxiliary_loss_clip": 0.01163588,
      "auxiliary_loss_mlp": 0.01022552,
      "balance_loss_clip": 0.97220999,
      "balance_loss_mlp": 1.01494074,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 1.900224284220436,
      "language_loss": 0.68360907,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70547044,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.675508499145508
    },
    {
      "auxiliary_loss_clip": 0.01180397,
      "auxiliary_loss_mlp": 0.0102694,
      "balance_loss_clip": 1.05421185,
      "balance_loss_mlp": 1.01904273,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 1.9945132451199694,
      "language_loss": 0.93588167,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.95795512,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.6070525646209717
    },
    {
      "auxiliary_loss_clip": 0.01176379,
      "auxiliary_loss_mlp": 0.01025479,
      "balance_loss_clip": 0.90027976,
      "balance_loss_mlp": 1.0180105,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 2.5787110082102727,
      "language_loss": 0.71316063,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73517919,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.7461137771606445
    },
    {
      "auxiliary_loss_clip": 0.01168789,
      "auxiliary_loss_mlp": 0.01122826,
      "balance_loss_clip": 0.97282833,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 1.8865955182023915,
      "language_loss": 0.86971796,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.89263415,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 2.7411842346191406
    },
    {
      "auxiliary_loss_clip": 0.01184071,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.01391768,
      "balance_loss_mlp": 1.02141333,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.2033987745399233,
      "language_loss": 0.82105446,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84319401,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.772435188293457
    },
    {
      "auxiliary_loss_clip": 0.01158958,
      "auxiliary_loss_mlp": 0.01024925,
      "balance_loss_clip": 0.89286077,
      "balance_loss_mlp": 1.01652706,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 2.0438417962990036,
      "language_loss": 0.72242415,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74426299,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.804072856903076
    },
    {
      "auxiliary_loss_clip": 0.01175776,
      "auxiliary_loss_mlp": 0.01032033,
      "balance_loss_clip": 1.0124011,
      "balance_loss_mlp": 1.02389669,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 2.333424879491109,
      "language_loss": 0.72590202,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.74798012,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.713942527770996
    },
    {
      "auxiliary_loss_clip": 0.01172238,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 0.93433338,
      "balance_loss_mlp": 1.01834941,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 1.7501963605301936,
      "language_loss": 0.72490686,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74689019,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.8717200756073
    },
    {
      "auxiliary_loss_clip": 0.01178218,
      "auxiliary_loss_mlp": 0.01025885,
      "balance_loss_clip": 0.93953723,
      "balance_loss_mlp": 1.01850033,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.6691924197788552,
      "language_loss": 0.76169789,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78373897,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 3.0347158908843994
    },
    {
      "auxiliary_loss_clip": 0.01180704,
      "auxiliary_loss_mlp": 0.01028684,
      "balance_loss_clip": 1.05438709,
      "balance_loss_mlp": 1.02076221,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 2.1019234710913643,
      "language_loss": 0.8516618,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.87375569,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 2.563976764678955
    },
    {
      "auxiliary_loss_clip": 0.01166117,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.01042032,
      "balance_loss_mlp": 1.02069306,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 2.4933251732386688,
      "language_loss": 0.69759655,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.7195369,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 2.543768882751465
    },
    {
      "auxiliary_loss_clip": 0.01172739,
      "auxiliary_loss_mlp": 0.01028901,
      "balance_loss_clip": 0.90010333,
      "balance_loss_mlp": 1.02145648,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 1.8704270337626112,
      "language_loss": 0.72457075,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.74658716,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 2.957411050796509
    },
    {
      "auxiliary_loss_clip": 0.01175257,
      "auxiliary_loss_mlp": 0.01023226,
      "balance_loss_clip": 1.05007708,
      "balance_loss_mlp": 1.01584959,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 2.789276558065326,
      "language_loss": 0.74772382,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.76970863,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 3.6269278526306152
    },
    {
      "auxiliary_loss_clip": 0.01183136,
      "auxiliary_loss_mlp": 0.01027799,
      "balance_loss_clip": 0.93784517,
      "balance_loss_mlp": 1.02074456,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 2.3886525553182194,
      "language_loss": 0.85217047,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87427986,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.7383079528808594
    },
    {
      "auxiliary_loss_clip": 0.01168711,
      "auxiliary_loss_mlp": 0.01123518,
      "balance_loss_clip": 0.93566012,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 1.5347495296629199,
      "language_loss": 0.78474623,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.80766845,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.680722951889038
    },
    {
      "auxiliary_loss_clip": 0.01091371,
      "auxiliary_loss_mlp": 0.00999692,
      "balance_loss_clip": 0.91310334,
      "balance_loss_mlp": 0.99796355,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7816928599084583,
      "language_loss": 0.59465575,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.61556637,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 4.355034589767456
    },
    {
      "auxiliary_loss_clip": 0.01170178,
      "auxiliary_loss_mlp": 0.01022988,
      "balance_loss_clip": 0.97304869,
      "balance_loss_mlp": 1.015293,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 2.0110455018985482,
      "language_loss": 0.82437193,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.84630358,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 2.69643235206604
    },
    {
      "auxiliary_loss_clip": 0.0118456,
      "auxiliary_loss_mlp": 0.01033014,
      "balance_loss_clip": 1.01962459,
      "balance_loss_mlp": 1.02493453,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 3.092673441619775,
      "language_loss": 0.79454607,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.8167218,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.624083995819092
    },
    {
      "auxiliary_loss_clip": 0.01158446,
      "auxiliary_loss_mlp": 0.01027608,
      "balance_loss_clip": 0.89643639,
      "balance_loss_mlp": 1.01973474,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.2372829952472872,
      "language_loss": 0.81484044,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.83670104,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 2.6822011470794678
    },
    {
      "auxiliary_loss_clip": 0.01181033,
      "auxiliary_loss_mlp": 0.01123364,
      "balance_loss_clip": 1.05366194,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 1.6317698267914824,
      "language_loss": 0.80802232,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.83106625,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 3.5818612575531006
    },
    {
      "auxiliary_loss_clip": 0.01158855,
      "auxiliary_loss_mlp": 0.01030218,
      "balance_loss_clip": 0.97084397,
      "balance_loss_mlp": 1.02240336,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 2.434741172168486,
      "language_loss": 0.69512427,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.71701509,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.7170064449310303
    },
    {
      "auxiliary_loss_clip": 0.01173163,
      "auxiliary_loss_mlp": 0.01033077,
      "balance_loss_clip": 0.93801016,
      "balance_loss_mlp": 1.02511966,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 1.720546853573058,
      "language_loss": 0.86185175,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88391411,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 2.7344837188720703
    },
    {
      "auxiliary_loss_clip": 0.01167584,
      "auxiliary_loss_mlp": 0.01026193,
      "balance_loss_clip": 0.97415829,
      "balance_loss_mlp": 1.01827168,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 1.662730923965574,
      "language_loss": 0.79936087,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.8212986,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 2.8685672283172607
    },
    {
      "auxiliary_loss_clip": 0.01181577,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.01782131,
      "balance_loss_mlp": 1.0173738,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 1.6249218739591074,
      "language_loss": 0.8109321,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.8329972,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 3.64652419090271
    },
    {
      "auxiliary_loss_clip": 0.01087262,
      "auxiliary_loss_mlp": 0.01006132,
      "balance_loss_clip": 0.99100769,
      "balance_loss_mlp": 1.00427198,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8652006369067357,
      "language_loss": 0.60650182,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.6274358,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.3811287879943848
    },
    {
      "auxiliary_loss_clip": 0.01180529,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.01410294,
      "balance_loss_mlp": 1.01922691,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 1.4998929289921235,
      "language_loss": 0.67456067,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69662976,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 3.0321006774902344
    },
    {
      "auxiliary_loss_clip": 0.01181006,
      "auxiliary_loss_mlp": 0.01022241,
      "balance_loss_clip": 1.01569271,
      "balance_loss_mlp": 1.01508224,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 2.492310964445858,
      "language_loss": 0.80399632,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.82602882,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.6933398246765137
    },
    {
      "auxiliary_loss_clip": 0.0118338,
      "auxiliary_loss_mlp": 0.01027206,
      "balance_loss_clip": 0.93629849,
      "balance_loss_mlp": 1.019207,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 3.8707993624890986,
      "language_loss": 0.81701618,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.83912206,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.724191904067993
    },
    {
      "auxiliary_loss_clip": 0.01179054,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.05302596,
      "balance_loss_mlp": 1.02246523,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.5941892535041438,
      "language_loss": 0.79091895,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.81301081,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.660287857055664
    },
    {
      "auxiliary_loss_clip": 0.01180407,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 1.05348158,
      "balance_loss_mlp": 1.01826429,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 2.133662448534502,
      "language_loss": 0.82076371,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.84282947,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 2.5753071308135986
    },
    {
      "auxiliary_loss_clip": 0.0117696,
      "auxiliary_loss_mlp": 0.01122569,
      "balance_loss_clip": 0.93994159,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 1.9113156457226155,
      "language_loss": 0.82838142,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.85137677,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.732783079147339
    },
    {
      "auxiliary_loss_clip": 0.01171996,
      "auxiliary_loss_mlp": 0.01025927,
      "balance_loss_clip": 0.97592509,
      "balance_loss_mlp": 1.01897085,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.5406718035790457,
      "language_loss": 0.83556551,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.85754472,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.7248618602752686
    },
    {
      "auxiliary_loss_clip": 0.01177062,
      "auxiliary_loss_mlp": 0.01030309,
      "balance_loss_clip": 1.05264175,
      "balance_loss_mlp": 1.02271521,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 2.501870196584648,
      "language_loss": 0.82864761,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.85072136,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.674142599105835
    },
    {
      "auxiliary_loss_clip": 0.01175015,
      "auxiliary_loss_mlp": 0.01122993,
      "balance_loss_clip": 0.93492991,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 1.477472397513293,
      "language_loss": 0.80619895,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.82917905,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.7383675575256348
    },
    {
      "auxiliary_loss_clip": 0.01166566,
      "auxiliary_loss_mlp": 0.01031211,
      "balance_loss_clip": 1.01114297,
      "balance_loss_mlp": 1.02322364,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 1.6576095478244668,
      "language_loss": 0.81021047,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.83218819,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 2.699814558029175
    },
    {
      "auxiliary_loss_clip": 0.01165578,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 0.86271572,
      "balance_loss_mlp": 1.02269697,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 1.890209048792157,
      "language_loss": 0.7199707,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74192882,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 2.7492213249206543
    },
    {
      "auxiliary_loss_clip": 0.01176858,
      "auxiliary_loss_mlp": 0.01024826,
      "balance_loss_clip": 1.01526785,
      "balance_loss_mlp": 1.01762581,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 3.545823873757292,
      "language_loss": 0.65523815,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67725492,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 2.7256462574005127
    },
    {
      "auxiliary_loss_clip": 0.01177279,
      "auxiliary_loss_mlp": 0.01030243,
      "balance_loss_clip": 0.97266304,
      "balance_loss_mlp": 1.02239299,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 2.24940457745536,
      "language_loss": 0.77112913,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79320443,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 2.733116865158081
    },
    {
      "auxiliary_loss_clip": 0.011553,
      "auxiliary_loss_mlp": 0.01027508,
      "balance_loss_clip": 0.93475264,
      "balance_loss_mlp": 1.0202837,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 1.5757638370734182,
      "language_loss": 0.78539431,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80722243,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 3.7279486656188965
    },
    {
      "auxiliary_loss_clip": 0.0118048,
      "auxiliary_loss_mlp": 0.01029619,
      "balance_loss_clip": 0.93636203,
      "balance_loss_mlp": 1.02204943,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 4.577215436667188,
      "language_loss": 0.75568867,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.77778965,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.7688474655151367
    },
    {
      "auxiliary_loss_clip": 0.01182614,
      "auxiliary_loss_mlp": 0.01031648,
      "balance_loss_clip": 1.01402807,
      "balance_loss_mlp": 1.02367306,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 2.2375645668596227,
      "language_loss": 0.68396622,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.70610881,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 3.6187164783477783
    },
    {
      "auxiliary_loss_clip": 0.01167441,
      "auxiliary_loss_mlp": 0.0112308,
      "balance_loss_clip": 0.89845902,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 1.6165067496927286,
      "language_loss": 0.63653052,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.65943575,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.872763156890869
    },
    {
      "auxiliary_loss_clip": 0.01085876,
      "auxiliary_loss_mlp": 0.01007317,
      "balance_loss_clip": 0.9149453,
      "balance_loss_mlp": 1.00537348,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6966657677299528,
      "language_loss": 0.52519292,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.54612482,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 3.3709123134613037
    },
    {
      "auxiliary_loss_clip": 0.01182499,
      "auxiliary_loss_mlp": 0.01029683,
      "balance_loss_clip": 0.93718481,
      "balance_loss_mlp": 1.02216101,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 1.7502323169315994,
      "language_loss": 0.72812986,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.75025165,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.8081305027008057
    },
    {
      "auxiliary_loss_clip": 0.01085924,
      "auxiliary_loss_mlp": 0.01005035,
      "balance_loss_clip": 0.90960675,
      "balance_loss_mlp": 1.0032109,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.8126602832668435,
      "language_loss": 0.55722177,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57813138,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.2566795349121094
    },
    {
      "auxiliary_loss_clip": 0.0118049,
      "auxiliary_loss_mlp": 0.01023735,
      "balance_loss_clip": 1.01332235,
      "balance_loss_mlp": 1.01630282,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.5984622930130652,
      "language_loss": 0.80709809,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.82914037,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 3.431626558303833
    },
    {
      "auxiliary_loss_clip": 0.01174751,
      "auxiliary_loss_mlp": 0.01033009,
      "balance_loss_clip": 1.01470923,
      "balance_loss_mlp": 1.02593958,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 1.9077496651667032,
      "language_loss": 0.88154721,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90362489,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.6721458435058594
    },
    {
      "auxiliary_loss_clip": 0.01177872,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 0.93672401,
      "balance_loss_mlp": 1.0230062,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.148995573669631,
      "language_loss": 0.80592799,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.82802415,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 2.700730085372925
    },
    {
      "auxiliary_loss_clip": 0.01173514,
      "auxiliary_loss_mlp": 0.01123161,
      "balance_loss_clip": 0.97444946,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.9964905930856405,
      "language_loss": 0.7704711,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.79343784,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 2.7576687335968018
    },
    {
      "auxiliary_loss_clip": 0.01163644,
      "auxiliary_loss_mlp": 0.01023556,
      "balance_loss_clip": 1.0105269,
      "balance_loss_mlp": 1.01557541,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 1.8668590325051946,
      "language_loss": 0.72093177,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.74280381,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 3.5544803142547607
    },
    {
      "auxiliary_loss_clip": 0.01181347,
      "auxiliary_loss_mlp": 0.01027073,
      "balance_loss_clip": 1.01424801,
      "balance_loss_mlp": 1.01975906,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 1.8109847690591856,
      "language_loss": 0.79553604,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.81762016,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.6171348094940186
    },
    {
      "auxiliary_loss_clip": 0.01174184,
      "auxiliary_loss_mlp": 0.01026323,
      "balance_loss_clip": 0.97550207,
      "balance_loss_mlp": 1.0189383,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 1.6473751831898273,
      "language_loss": 0.7765801,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.79858518,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.708477735519409
    },
    {
      "auxiliary_loss_clip": 0.01175221,
      "auxiliary_loss_mlp": 0.01028974,
      "balance_loss_clip": 0.97358888,
      "balance_loss_mlp": 1.02132106,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 1.6977638307471823,
      "language_loss": 0.78564858,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.80769056,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.7168350219726562
    },
    {
      "auxiliary_loss_clip": 0.0116953,
      "auxiliary_loss_mlp": 0.01029852,
      "balance_loss_clip": 0.89644337,
      "balance_loss_mlp": 1.02187657,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.1503138058911753,
      "language_loss": 0.73989093,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.76188475,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 2.678866386413574
    },
    {
      "auxiliary_loss_clip": 0.01175568,
      "auxiliary_loss_mlp": 0.01024404,
      "balance_loss_clip": 0.97569513,
      "balance_loss_mlp": 1.01723051,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.4332653609072659,
      "language_loss": 0.81614363,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.83814335,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.724531650543213
    },
    {
      "auxiliary_loss_clip": 0.0117017,
      "auxiliary_loss_mlp": 0.01026667,
      "balance_loss_clip": 0.93819243,
      "balance_loss_mlp": 1.0186857,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 1.9879200875964933,
      "language_loss": 0.6982283,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.72019666,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 2.7473948001861572
    },
    {
      "auxiliary_loss_clip": 0.01178108,
      "auxiliary_loss_mlp": 0.01029398,
      "balance_loss_clip": 1.05282736,
      "balance_loss_mlp": 1.02147079,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 1.9823999554207854,
      "language_loss": 0.71487194,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.736947,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.6388959884643555
    },
    {
      "auxiliary_loss_clip": 0.01175116,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 1.01671648,
      "balance_loss_mlp": 1.01905298,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 1.851173965644763,
      "language_loss": 0.81941247,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.84142935,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.693971633911133
    },
    {
      "auxiliary_loss_clip": 0.01160612,
      "auxiliary_loss_mlp": 0.0102971,
      "balance_loss_clip": 0.93297637,
      "balance_loss_mlp": 1.02171659,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.162418973242262,
      "language_loss": 0.72587931,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.74778253,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.7221286296844482
    },
    {
      "auxiliary_loss_clip": 0.0117766,
      "auxiliary_loss_mlp": 0.0103041,
      "balance_loss_clip": 0.97780704,
      "balance_loss_mlp": 1.02258396,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 1.9445508975196728,
      "language_loss": 0.78273696,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80481762,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.6464879512786865
    },
    {
      "auxiliary_loss_clip": 0.01174026,
      "auxiliary_loss_mlp": 0.01032913,
      "balance_loss_clip": 0.9779619,
      "balance_loss_mlp": 1.02566564,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 1.7422697013122663,
      "language_loss": 0.88014793,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90221727,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.729095697402954
    },
    {
      "auxiliary_loss_clip": 0.01083957,
      "auxiliary_loss_mlp": 0.00999438,
      "balance_loss_clip": 0.98546898,
      "balance_loss_mlp": 0.99761409,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.6992301399397682,
      "language_loss": 0.57602614,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59686011,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 3.3852412700653076
    },
    {
      "auxiliary_loss_clip": 0.01174652,
      "auxiliary_loss_mlp": 0.01030234,
      "balance_loss_clip": 1.01199579,
      "balance_loss_mlp": 1.0227356,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.4866674331594936,
      "language_loss": 0.83050108,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.85254991,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 2.691603183746338
    },
    {
      "auxiliary_loss_clip": 0.01177613,
      "auxiliary_loss_mlp": 0.01031039,
      "balance_loss_clip": 1.01294231,
      "balance_loss_mlp": 1.02310562,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 2.624477682109535,
      "language_loss": 0.81035036,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83243686,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.6597719192504883
    },
    {
      "auxiliary_loss_clip": 0.01151732,
      "auxiliary_loss_mlp": 0.01122363,
      "balance_loss_clip": 0.89388585,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 1.908344528532374,
      "language_loss": 0.80180949,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.82455039,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 3.7005317211151123
    },
    {
      "auxiliary_loss_clip": 0.01175346,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.05165637,
      "balance_loss_mlp": 1.01742983,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 3.148780622872174,
      "language_loss": 0.78241491,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.80441594,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.6173248291015625
    },
    {
      "auxiliary_loss_clip": 0.01079905,
      "auxiliary_loss_mlp": 0.01000428,
      "balance_loss_clip": 0.87094933,
      "balance_loss_mlp": 0.99858016,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.8363103168153869,
      "language_loss": 0.57395673,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59476006,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 4.42167329788208
    },
    {
      "auxiliary_loss_clip": 0.0117875,
      "auxiliary_loss_mlp": 0.01028906,
      "balance_loss_clip": 0.89700484,
      "balance_loss_mlp": 1.02038872,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 1.9270239576933657,
      "language_loss": 0.66326261,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68533915,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.6644034385681152
    },
    {
      "auxiliary_loss_clip": 0.0116905,
      "auxiliary_loss_mlp": 0.01031268,
      "balance_loss_clip": 0.93138134,
      "balance_loss_mlp": 1.0233047,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 4.166307305918473,
      "language_loss": 0.6621474,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.68415058,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.7103097438812256
    },
    {
      "auxiliary_loss_clip": 0.01184942,
      "auxiliary_loss_mlp": 0.01032922,
      "balance_loss_clip": 0.8998462,
      "balance_loss_mlp": 1.02479196,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 2.3961046165310926,
      "language_loss": 0.65536958,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.67754817,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 2.7786288261413574
    },
    {
      "auxiliary_loss_clip": 0.01175347,
      "auxiliary_loss_mlp": 0.01028054,
      "balance_loss_clip": 0.97272414,
      "balance_loss_mlp": 1.0203923,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 2.1655280716905008,
      "language_loss": 0.76341796,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78545201,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.6828012466430664
    },
    {
      "auxiliary_loss_clip": 0.01173408,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 0.97599816,
      "balance_loss_mlp": 1.0209446,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 1.8178484624294768,
      "language_loss": 0.76446998,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.78648877,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 3.5592596530914307
    },
    {
      "auxiliary_loss_clip": 0.01169368,
      "auxiliary_loss_mlp": 0.01123265,
      "balance_loss_clip": 0.93817407,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 2.2445874040393567,
      "language_loss": 0.60154045,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.62446678,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.685035467147827
    },
    {
      "auxiliary_loss_clip": 0.0116898,
      "auxiliary_loss_mlp": 0.01030532,
      "balance_loss_clip": 0.97389507,
      "balance_loss_mlp": 1.02268481,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 2.1831836503261637,
      "language_loss": 0.85677719,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.87877232,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.668843984603882
    },
    {
      "auxiliary_loss_clip": 0.01172275,
      "auxiliary_loss_mlp": 0.01026848,
      "balance_loss_clip": 1.01353455,
      "balance_loss_mlp": 1.01976132,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 1.6428684917644099,
      "language_loss": 0.79186594,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.8138572,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 2.614903450012207
    },
    {
      "auxiliary_loss_clip": 0.01175102,
      "auxiliary_loss_mlp": 0.01033396,
      "balance_loss_clip": 0.97281533,
      "balance_loss_mlp": 1.02567744,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 1.8118306321324298,
      "language_loss": 0.82981205,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.85189706,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 3.592318534851074
    },
    {
      "auxiliary_loss_clip": 0.01180852,
      "auxiliary_loss_mlp": 0.01038353,
      "balance_loss_clip": 0.97555566,
      "balance_loss_mlp": 1.03020513,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 2.0408262513622804,
      "language_loss": 0.74435914,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.76655126,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.7152206897735596
    },
    {
      "auxiliary_loss_clip": 0.01181081,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 0.93635517,
      "balance_loss_mlp": 1.01980615,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 2.558769104060844,
      "language_loss": 0.81105322,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.83314002,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.6952834129333496
    },
    {
      "auxiliary_loss_clip": 0.01176372,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 0.97704661,
      "balance_loss_mlp": 1.01914847,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 1.7900384000776806,
      "language_loss": 0.73799312,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76002729,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.648247718811035
    },
    {
      "auxiliary_loss_clip": 0.01175215,
      "auxiliary_loss_mlp": 0.01028133,
      "balance_loss_clip": 0.94011486,
      "balance_loss_mlp": 1.01997316,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 1.8395969246548842,
      "language_loss": 0.75548065,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.77751422,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 2.7433745861053467
    },
    {
      "auxiliary_loss_clip": 0.01176657,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.01463425,
      "balance_loss_mlp": 1.01695883,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 1.7414685360590882,
      "language_loss": 0.74299169,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76500154,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 2.711043357849121
    },
    {
      "auxiliary_loss_clip": 0.01179146,
      "auxiliary_loss_mlp": 0.01033545,
      "balance_loss_clip": 1.01602113,
      "balance_loss_mlp": 1.02555192,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 1.9645027710471392,
      "language_loss": 0.71751535,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.73964226,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.715564727783203
    },
    {
      "auxiliary_loss_clip": 0.01169461,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 0.93785256,
      "balance_loss_mlp": 1.01944828,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.054212691711671,
      "language_loss": 0.89672887,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.91869974,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.7609055042266846
    },
    {
      "auxiliary_loss_clip": 0.01163148,
      "auxiliary_loss_mlp": 0.01027088,
      "balance_loss_clip": 0.90012777,
      "balance_loss_mlp": 1.01967287,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 1.8432894787470195,
      "language_loss": 0.9031117,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92501402,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.7155137062072754
    },
    {
      "auxiliary_loss_clip": 0.01168141,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 0.93777359,
      "balance_loss_mlp": 1.02298427,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 1.828544831038968,
      "language_loss": 0.80319619,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82518113,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.7884879112243652
    },
    {
      "auxiliary_loss_clip": 0.01173341,
      "auxiliary_loss_mlp": 0.0103037,
      "balance_loss_clip": 0.94103408,
      "balance_loss_mlp": 1.02235937,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 2.48591924067585,
      "language_loss": 0.78615814,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.80819523,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.687934398651123
    },
    {
      "auxiliary_loss_clip": 0.01173784,
      "auxiliary_loss_mlp": 0.01024595,
      "balance_loss_clip": 1.01390529,
      "balance_loss_mlp": 1.01735866,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 2.013711349618487,
      "language_loss": 0.80738819,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.82937193,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 2.6196625232696533
    },
    {
      "auxiliary_loss_clip": 0.01176629,
      "auxiliary_loss_mlp": 0.01028827,
      "balance_loss_clip": 1.01137245,
      "balance_loss_mlp": 1.02091169,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.4099116289140015,
      "language_loss": 0.69336867,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.71542317,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 2.6424007415771484
    },
    {
      "auxiliary_loss_clip": 0.0117667,
      "auxiliary_loss_mlp": 0.01026393,
      "balance_loss_clip": 1.01434445,
      "balance_loss_mlp": 1.0191803,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 2.2069108478402426,
      "language_loss": 0.68999779,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.71202838,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 2.6471378803253174
    },
    {
      "auxiliary_loss_clip": 0.01177895,
      "auxiliary_loss_mlp": 0.01033638,
      "balance_loss_clip": 1.01658249,
      "balance_loss_mlp": 1.0257287,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 1.7201117217599398,
      "language_loss": 0.69907081,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72118616,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 2.6249139308929443
    },
    {
      "auxiliary_loss_clip": 0.01172474,
      "auxiliary_loss_mlp": 0.01022311,
      "balance_loss_clip": 0.97744048,
      "balance_loss_mlp": 1.01516449,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.9315112342228633,
      "language_loss": 0.80622214,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82817,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 3.716747283935547
    },
    {
      "auxiliary_loss_clip": 0.01177834,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 1.05092669,
      "balance_loss_mlp": 1.01918399,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 1.826688187660483,
      "language_loss": 0.6808064,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70285594,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.7014689445495605
    },
    {
      "auxiliary_loss_clip": 0.01180919,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.05380106,
      "balance_loss_mlp": 1.02178359,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 2.353755405447872,
      "language_loss": 0.66470891,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68681461,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 3.5954954624176025
    },
    {
      "auxiliary_loss_clip": 0.01181455,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 0.97483146,
      "balance_loss_mlp": 1.02176929,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 1.6524117745300972,
      "language_loss": 0.76792443,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.79004222,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.691923141479492
    },
    {
      "auxiliary_loss_clip": 0.01184866,
      "auxiliary_loss_mlp": 0.01032949,
      "balance_loss_clip": 1.01643026,
      "balance_loss_mlp": 1.02528119,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 1.6712849345441911,
      "language_loss": 0.76369488,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.785873,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.6356945037841797
    },
    {
      "auxiliary_loss_clip": 0.01173018,
      "auxiliary_loss_mlp": 0.01122987,
      "balance_loss_clip": 1.01332927,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.2538434120017263,
      "language_loss": 0.70821404,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.73117411,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.636291980743408
    },
    {
      "auxiliary_loss_clip": 0.01169948,
      "auxiliary_loss_mlp": 0.01029697,
      "balance_loss_clip": 0.97410244,
      "balance_loss_mlp": 1.02202582,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.394966342243936,
      "language_loss": 0.78382587,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80582237,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.7753961086273193
    },
    {
      "auxiliary_loss_clip": 0.01172836,
      "auxiliary_loss_mlp": 0.01022046,
      "balance_loss_clip": 0.93450326,
      "balance_loss_mlp": 1.01503026,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.5099486811330385,
      "language_loss": 0.66507679,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.68702561,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 3.692641496658325
    },
    {
      "auxiliary_loss_clip": 0.0117131,
      "auxiliary_loss_mlp": 0.01033562,
      "balance_loss_clip": 0.97443563,
      "balance_loss_mlp": 1.02544963,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 1.6168730527496507,
      "language_loss": 0.82477069,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.8468194,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 2.6586952209472656
    },
    {
      "auxiliary_loss_clip": 0.01171429,
      "auxiliary_loss_mlp": 0.01025314,
      "balance_loss_clip": 1.01325905,
      "balance_loss_mlp": 1.01800668,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 1.6161051086602554,
      "language_loss": 0.83852983,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86049724,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.6863973140716553
    },
    {
      "auxiliary_loss_clip": 0.0117627,
      "auxiliary_loss_mlp": 0.01025349,
      "balance_loss_clip": 0.97648561,
      "balance_loss_mlp": 1.01776147,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 1.665409622005013,
      "language_loss": 0.62859511,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65061128,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 2.701368570327759
    },
    {
      "auxiliary_loss_clip": 0.01168718,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 0.9734621,
      "balance_loss_mlp": 1.0196979,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 1.8842277146101687,
      "language_loss": 0.71217883,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.73413706,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 3.6980140209198
    },
    {
      "auxiliary_loss_clip": 0.01160495,
      "auxiliary_loss_mlp": 0.01029184,
      "balance_loss_clip": 0.93358171,
      "balance_loss_mlp": 1.02164412,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 2.220741914965985,
      "language_loss": 0.74000823,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.76190507,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 2.751681327819824
    },
    {
      "auxiliary_loss_clip": 0.01179552,
      "auxiliary_loss_mlp": 0.01025641,
      "balance_loss_clip": 1.0524348,
      "balance_loss_mlp": 1.01818132,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 1.7912562162717853,
      "language_loss": 0.75893962,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.78099149,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.647508144378662
    },
    {
      "auxiliary_loss_clip": 0.01167516,
      "auxiliary_loss_mlp": 0.01026611,
      "balance_loss_clip": 0.97563338,
      "balance_loss_mlp": 1.01940489,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.579641845253382,
      "language_loss": 0.8064574,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.8283987,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.679215908050537
    },
    {
      "auxiliary_loss_clip": 0.011681,
      "auxiliary_loss_mlp": 0.01032599,
      "balance_loss_clip": 1.01067889,
      "balance_loss_mlp": 1.02520227,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 2.0824439032533473,
      "language_loss": 0.65839958,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.68040657,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.608609676361084
    },
    {
      "auxiliary_loss_clip": 0.01178347,
      "auxiliary_loss_mlp": 0.01021289,
      "balance_loss_clip": 0.97782391,
      "balance_loss_mlp": 1.01357317,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 1.789181470778224,
      "language_loss": 0.74841893,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.77041531,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 2.7071080207824707
    },
    {
      "auxiliary_loss_clip": 0.01177948,
      "auxiliary_loss_mlp": 0.01030732,
      "balance_loss_clip": 1.0160445,
      "balance_loss_mlp": 1.02305496,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 1.9685151854281044,
      "language_loss": 0.86092621,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88301301,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.6317763328552246
    },
    {
      "auxiliary_loss_clip": 0.01170556,
      "auxiliary_loss_mlp": 0.01029303,
      "balance_loss_clip": 1.01021385,
      "balance_loss_mlp": 1.02140522,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 1.6263211670249968,
      "language_loss": 0.71124458,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73324317,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.663574457168579
    },
    {
      "auxiliary_loss_clip": 0.01168395,
      "auxiliary_loss_mlp": 0.01027859,
      "balance_loss_clip": 0.89572686,
      "balance_loss_mlp": 1.01853085,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 3.036208784246846,
      "language_loss": 0.75096673,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77292925,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.8307552337646484
    },
    {
      "auxiliary_loss_clip": 0.01168005,
      "auxiliary_loss_mlp": 0.01037544,
      "balance_loss_clip": 0.89885312,
      "balance_loss_mlp": 1.02982545,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.0062662756743466,
      "language_loss": 0.77238393,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.79443949,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.746704578399658
    },
    {
      "auxiliary_loss_clip": 0.01183457,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 1.01641083,
      "balance_loss_mlp": 1.02183306,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 3.3368385333289727,
      "language_loss": 0.49104625,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.51317894,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.722280502319336
    },
    {
      "auxiliary_loss_clip": 0.01175792,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.01068091,
      "balance_loss_mlp": 1.02017558,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 1.8414940589769178,
      "language_loss": 0.77876973,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.80080962,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 2.6505913734436035
    },
    {
      "auxiliary_loss_clip": 0.01167764,
      "auxiliary_loss_mlp": 0.01028918,
      "balance_loss_clip": 0.97588217,
      "balance_loss_mlp": 1.02156329,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 1.62617777540037,
      "language_loss": 0.752783,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.77474982,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 2.6851227283477783
    },
    {
      "auxiliary_loss_clip": 0.01081192,
      "auxiliary_loss_mlp": 0.01005515,
      "balance_loss_clip": 0.94600058,
      "balance_loss_mlp": 1.00352383,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.8787276460709612,
      "language_loss": 0.62696636,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64783335,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 3.149780511856079
    },
    {
      "auxiliary_loss_clip": 0.01175781,
      "auxiliary_loss_mlp": 0.01028647,
      "balance_loss_clip": 1.05294263,
      "balance_loss_mlp": 1.02155077,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 1.8970123561321226,
      "language_loss": 0.69514215,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71718633,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.599177122116089
    },
    {
      "auxiliary_loss_clip": 0.01086933,
      "auxiliary_loss_mlp": 0.01001722,
      "balance_loss_clip": 0.87223935,
      "balance_loss_mlp": 0.99989861,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.8845015032371883,
      "language_loss": 0.66868472,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68957126,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 4.186995267868042
    },
    {
      "auxiliary_loss_clip": 0.01173311,
      "auxiliary_loss_mlp": 0.01024425,
      "balance_loss_clip": 1.05117226,
      "balance_loss_mlp": 1.01698935,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.7163393903090887,
      "language_loss": 0.72297108,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.74494845,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.7198355197906494
    },
    {
      "auxiliary_loss_clip": 0.01169368,
      "auxiliary_loss_mlp": 0.0102481,
      "balance_loss_clip": 0.92979395,
      "balance_loss_mlp": 1.01645398,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.0165397903499116,
      "language_loss": 0.64819694,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.67013872,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 3.8033974170684814
    },
    {
      "auxiliary_loss_clip": 0.01171206,
      "auxiliary_loss_mlp": 0.01026577,
      "balance_loss_clip": 0.93522382,
      "balance_loss_mlp": 1.01897156,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 1.598229979483385,
      "language_loss": 0.85120016,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87317801,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.7158267498016357
    },
    {
      "auxiliary_loss_clip": 0.01181307,
      "auxiliary_loss_mlp": 0.01027061,
      "balance_loss_clip": 0.97495675,
      "balance_loss_mlp": 1.01926446,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 2.0718413144486236,
      "language_loss": 0.80164158,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82372522,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.704508066177368
    },
    {
      "auxiliary_loss_clip": 0.01174545,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 0.97416091,
      "balance_loss_mlp": 1.02158999,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 1.7419364518791767,
      "language_loss": 0.82514322,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.84717476,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.6315648555755615
    },
    {
      "auxiliary_loss_clip": 0.01164553,
      "auxiliary_loss_mlp": 0.01032084,
      "balance_loss_clip": 0.93083,
      "balance_loss_mlp": 1.02437091,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 1.8269074431498737,
      "language_loss": 0.73033959,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.75230598,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.765068769454956
    },
    {
      "auxiliary_loss_clip": 0.01169817,
      "auxiliary_loss_mlp": 0.01029693,
      "balance_loss_clip": 1.01138234,
      "balance_loss_mlp": 1.02213502,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.150018793319201,
      "language_loss": 0.84320027,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86519533,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 3.5218727588653564
    },
    {
      "auxiliary_loss_clip": 0.01179541,
      "auxiliary_loss_mlp": 0.01029778,
      "balance_loss_clip": 0.97384655,
      "balance_loss_mlp": 1.02174377,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 2.4103546655654475,
      "language_loss": 0.59229159,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.61438477,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 2.5864953994750977
    },
    {
      "auxiliary_loss_clip": 0.01175229,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 0.89732563,
      "balance_loss_mlp": 1.01847625,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 1.7485331732736837,
      "language_loss": 0.82656211,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.84857786,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.708543062210083
    },
    {
      "auxiliary_loss_clip": 0.01177863,
      "auxiliary_loss_mlp": 0.01026217,
      "balance_loss_clip": 1.01293874,
      "balance_loss_mlp": 1.01813424,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 1.5956796927393602,
      "language_loss": 0.84801495,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.87005579,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.718959093093872
    },
    {
      "auxiliary_loss_clip": 0.01173315,
      "auxiliary_loss_mlp": 0.01023912,
      "balance_loss_clip": 0.97447741,
      "balance_loss_mlp": 1.0161097,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.4393580112642654,
      "language_loss": 0.73249108,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75446337,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 3.6376760005950928
    },
    {
      "auxiliary_loss_clip": 0.01167989,
      "auxiliary_loss_mlp": 0.01026403,
      "balance_loss_clip": 0.97320622,
      "balance_loss_mlp": 1.01865387,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.1417142749210525,
      "language_loss": 0.67316127,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.6951052,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.812887668609619
    },
    {
      "auxiliary_loss_clip": 0.01169482,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.01085711,
      "balance_loss_mlp": 1.01881313,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 1.9979703631799874,
      "language_loss": 0.77013886,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79210615,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.6616435050964355
    },
    {
      "auxiliary_loss_clip": 0.01186108,
      "auxiliary_loss_mlp": 0.01027739,
      "balance_loss_clip": 0.93971419,
      "balance_loss_mlp": 1.01910233,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 2.3700497813876975,
      "language_loss": 0.8180089,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.84014738,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.661534309387207
    },
    {
      "auxiliary_loss_clip": 0.01173269,
      "auxiliary_loss_mlp": 0.01030076,
      "balance_loss_clip": 0.97388035,
      "balance_loss_mlp": 1.02204132,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 2.0066686961895646,
      "language_loss": 0.66247374,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68450719,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.751826047897339
    },
    {
      "auxiliary_loss_clip": 0.01162282,
      "auxiliary_loss_mlp": 0.01026886,
      "balance_loss_clip": 0.93433475,
      "balance_loss_mlp": 1.01916122,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.1717181692746497,
      "language_loss": 0.81304461,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83493626,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 2.701000213623047
    },
    {
      "auxiliary_loss_clip": 0.0117595,
      "auxiliary_loss_mlp": 0.01024989,
      "balance_loss_clip": 1.05335665,
      "balance_loss_mlp": 1.01715684,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 2.3151423272870635,
      "language_loss": 0.82066929,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.84267873,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 2.6536948680877686
    },
    {
      "auxiliary_loss_clip": 0.01179903,
      "auxiliary_loss_mlp": 0.0102784,
      "balance_loss_clip": 1.05443513,
      "balance_loss_mlp": 1.02019322,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.1273178822269414,
      "language_loss": 0.78637171,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.80844915,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.6552765369415283
    },
    {
      "auxiliary_loss_clip": 0.01158302,
      "auxiliary_loss_mlp": 0.01024378,
      "balance_loss_clip": 0.89493561,
      "balance_loss_mlp": 1.01730299,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 1.6973496839259974,
      "language_loss": 0.76025784,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78208458,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.7786037921905518
    },
    {
      "auxiliary_loss_clip": 0.01172126,
      "auxiliary_loss_mlp": 0.01025035,
      "balance_loss_clip": 0.97435558,
      "balance_loss_mlp": 1.01740003,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 1.7736945543432674,
      "language_loss": 0.76413167,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.78610331,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 2.6575300693511963
    },
    {
      "auxiliary_loss_clip": 0.0117336,
      "auxiliary_loss_mlp": 0.01019508,
      "balance_loss_clip": 1.01522732,
      "balance_loss_mlp": 1.01259375,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.591752877129162,
      "language_loss": 0.81498945,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83691812,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.606964588165283
    },
    {
      "auxiliary_loss_clip": 0.01173471,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.01363468,
      "balance_loss_mlp": 1.02144563,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 1.5299684417867614,
      "language_loss": 0.82303667,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.845061,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 2.6692404747009277
    },
    {
      "auxiliary_loss_clip": 0.01178262,
      "auxiliary_loss_mlp": 0.01026516,
      "balance_loss_clip": 1.01317883,
      "balance_loss_mlp": 1.01921415,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.3450865264130214,
      "language_loss": 0.80970395,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.8317517,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 2.6275620460510254
    },
    {
      "auxiliary_loss_clip": 0.01163757,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 0.93399847,
      "balance_loss_mlp": 1.01882732,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.5376816477748827,
      "language_loss": 0.69350898,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.71541888,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 2.7269716262817383
    },
    {
      "auxiliary_loss_clip": 0.01087942,
      "auxiliary_loss_mlp": 0.01006835,
      "balance_loss_clip": 0.83253086,
      "balance_loss_mlp": 1.00489235,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7504317705183794,
      "language_loss": 0.55467606,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57562387,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.50167179107666
    },
    {
      "auxiliary_loss_clip": 0.0116634,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.0098896,
      "balance_loss_mlp": 1.0192256,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.103062870419619,
      "language_loss": 0.71643221,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.73836529,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 4.250727891921997
    },
    {
      "auxiliary_loss_clip": 0.01171692,
      "auxiliary_loss_mlp": 0.01023529,
      "balance_loss_clip": 0.97671998,
      "balance_loss_mlp": 1.01604283,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 1.6463690141920666,
      "language_loss": 0.67946273,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70141494,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.7525479793548584
    },
    {
      "auxiliary_loss_clip": 0.01161886,
      "auxiliary_loss_mlp": 0.01026731,
      "balance_loss_clip": 0.93163681,
      "balance_loss_mlp": 1.01913154,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 2.637008200999252,
      "language_loss": 0.81723022,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.83911633,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 3.710582733154297
    },
    {
      "auxiliary_loss_clip": 0.01172957,
      "auxiliary_loss_mlp": 0.01031634,
      "balance_loss_clip": 0.97720242,
      "balance_loss_mlp": 1.0232892,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.4429533431432957,
      "language_loss": 0.78880787,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81085384,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.7541894912719727
    },
    {
      "auxiliary_loss_clip": 0.01174431,
      "auxiliary_loss_mlp": 0.01030229,
      "balance_loss_clip": 0.9356972,
      "balance_loss_mlp": 1.02241206,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 1.8531688950821659,
      "language_loss": 0.82185376,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84390038,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.777848482131958
    },
    {
      "auxiliary_loss_clip": 0.01176471,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.01285434,
      "balance_loss_mlp": 1.02078843,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 1.9802483612733106,
      "language_loss": 0.74175858,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76380497,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.5945327281951904
    },
    {
      "auxiliary_loss_clip": 0.01179915,
      "auxiliary_loss_mlp": 0.01032544,
      "balance_loss_clip": 1.01647472,
      "balance_loss_mlp": 1.02506328,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.8370864886285694,
      "language_loss": 0.76946628,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.79159087,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.714538812637329
    },
    {
      "auxiliary_loss_clip": 0.01171131,
      "auxiliary_loss_mlp": 0.01030708,
      "balance_loss_clip": 0.93436909,
      "balance_loss_mlp": 1.02295935,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 1.9271447158093122,
      "language_loss": 0.82670283,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.84872127,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 3.7060344219207764
    },
    {
      "auxiliary_loss_clip": 0.01169768,
      "auxiliary_loss_mlp": 0.01023351,
      "balance_loss_clip": 1.01102948,
      "balance_loss_mlp": 1.01600766,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.152863118482644,
      "language_loss": 0.84280324,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.86473441,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.5889241695404053
    },
    {
      "auxiliary_loss_clip": 0.01085598,
      "auxiliary_loss_mlp": 0.01117091,
      "balance_loss_clip": 0.91004539,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.9071726969037395,
      "language_loss": 0.55670071,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.5787276,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.296612024307251
    },
    {
      "auxiliary_loss_clip": 0.01173907,
      "auxiliary_loss_mlp": 0.01122557,
      "balance_loss_clip": 0.97772682,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 1.840848427207644,
      "language_loss": 0.83262831,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.85559291,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.6744742393493652
    },
    {
      "auxiliary_loss_clip": 0.01171513,
      "auxiliary_loss_mlp": 0.01123341,
      "balance_loss_clip": 1.01387334,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 1.6360756433955022,
      "language_loss": 0.77022254,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.79317105,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 3.605560779571533
    },
    {
      "auxiliary_loss_clip": 0.01178559,
      "auxiliary_loss_mlp": 0.01029718,
      "balance_loss_clip": 0.89912772,
      "balance_loss_mlp": 1.0218327,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 1.9425275677635285,
      "language_loss": 0.70256752,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72465032,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 2.839046001434326
    },
    {
      "auxiliary_loss_clip": 0.01174951,
      "auxiliary_loss_mlp": 0.01030415,
      "balance_loss_clip": 1.0112133,
      "balance_loss_mlp": 1.02199841,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 2.1966245736423313,
      "language_loss": 0.88917267,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.91122639,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 2.639350414276123
    },
    {
      "auxiliary_loss_clip": 0.01177165,
      "auxiliary_loss_mlp": 0.01029522,
      "balance_loss_clip": 1.01489079,
      "balance_loss_mlp": 1.02240205,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 1.676958029092068,
      "language_loss": 0.80174971,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.8238166,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.713242530822754
    },
    {
      "auxiliary_loss_clip": 0.01168661,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 0.97544897,
      "balance_loss_mlp": 1.02105474,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.6633842457422745,
      "language_loss": 0.80160695,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.8235811,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.695765256881714
    },
    {
      "auxiliary_loss_clip": 0.01173027,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 0.93768257,
      "balance_loss_mlp": 1.01787448,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 3.176450576808386,
      "language_loss": 0.70824409,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.73022622,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.776684045791626
    },
    {
      "auxiliary_loss_clip": 0.01171423,
      "auxiliary_loss_mlp": 0.01028511,
      "balance_loss_clip": 0.97695625,
      "balance_loss_mlp": 1.02161217,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 1.6427562226039394,
      "language_loss": 0.84847617,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.87047553,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 2.710658073425293
    },
    {
      "auxiliary_loss_clip": 0.01171888,
      "auxiliary_loss_mlp": 0.0102689,
      "balance_loss_clip": 0.93420303,
      "balance_loss_mlp": 1.01938629,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 1.789641602294654,
      "language_loss": 0.83307266,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85506046,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.7624974250793457
    },
    {
      "auxiliary_loss_clip": 0.01174829,
      "auxiliary_loss_mlp": 0.01031202,
      "balance_loss_clip": 0.93553853,
      "balance_loss_mlp": 1.02335835,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 2.10594424872241,
      "language_loss": 0.86874926,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89080954,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.6979758739471436
    },
    {
      "auxiliary_loss_clip": 0.01181068,
      "auxiliary_loss_mlp": 0.01023856,
      "balance_loss_clip": 0.97885776,
      "balance_loss_mlp": 1.01626277,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 1.6991158926405947,
      "language_loss": 0.84253508,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86458433,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 2.7090795040130615
    },
    {
      "auxiliary_loss_clip": 0.01161536,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 0.93357909,
      "balance_loss_mlp": 1.0191927,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.6984307152906624,
      "language_loss": 0.75718689,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.7790643,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.674281120300293
    },
    {
      "auxiliary_loss_clip": 0.01170793,
      "auxiliary_loss_mlp": 0.01028966,
      "balance_loss_clip": 0.93582606,
      "balance_loss_mlp": 1.02127385,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 1.6103859881465232,
      "language_loss": 0.74730062,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.7692982,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 2.747267246246338
    },
    {
      "auxiliary_loss_clip": 0.01169138,
      "auxiliary_loss_mlp": 0.01029918,
      "balance_loss_clip": 1.01057601,
      "balance_loss_mlp": 1.02140033,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 2.12368133441766,
      "language_loss": 0.83117044,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85316098,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 2.6630005836486816
    },
    {
      "auxiliary_loss_clip": 0.01161415,
      "auxiliary_loss_mlp": 0.01123174,
      "balance_loss_clip": 0.96974027,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 1.9169332046129193,
      "language_loss": 0.71933746,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.74218339,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 2.6739139556884766
    },
    {
      "auxiliary_loss_clip": 0.01172268,
      "auxiliary_loss_mlp": 0.01031397,
      "balance_loss_clip": 0.97754169,
      "balance_loss_mlp": 1.02366614,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 3.644257911198776,
      "language_loss": 0.7338289,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75586557,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 2.73640775680542
    },
    {
      "auxiliary_loss_clip": 0.0117493,
      "auxiliary_loss_mlp": 0.01122829,
      "balance_loss_clip": 0.97335052,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 1.5783140676029233,
      "language_loss": 0.75063956,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.77361715,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 3.744321584701538
    },
    {
      "auxiliary_loss_clip": 0.01178446,
      "auxiliary_loss_mlp": 0.01029515,
      "balance_loss_clip": 0.97583413,
      "balance_loss_mlp": 1.0214895,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 1.8723428195365568,
      "language_loss": 0.84549654,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.86757618,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.793437957763672
    },
    {
      "auxiliary_loss_clip": 0.01174441,
      "auxiliary_loss_mlp": 0.01025656,
      "balance_loss_clip": 0.97316426,
      "balance_loss_mlp": 1.01810122,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 1.9146957744479902,
      "language_loss": 0.77879608,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.80079705,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.6838014125823975
    },
    {
      "auxiliary_loss_clip": 0.01161004,
      "auxiliary_loss_mlp": 0.01030293,
      "balance_loss_clip": 0.96808541,
      "balance_loss_mlp": 1.02244329,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 1.6232010865671822,
      "language_loss": 0.69626021,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.71817327,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 3.6936161518096924
    },
    {
      "auxiliary_loss_clip": 0.01171506,
      "auxiliary_loss_mlp": 0.01023202,
      "balance_loss_clip": 1.04823196,
      "balance_loss_mlp": 1.01587641,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.6232011311579748,
      "language_loss": 0.73805726,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.76000428,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.6641740798950195
    },
    {
      "auxiliary_loss_clip": 0.01171623,
      "auxiliary_loss_mlp": 0.01030029,
      "balance_loss_clip": 0.9703657,
      "balance_loss_mlp": 1.02191651,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 1.719522224823814,
      "language_loss": 0.63594377,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.6579603,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.755009651184082
    },
    {
      "auxiliary_loss_clip": 0.01179373,
      "auxiliary_loss_mlp": 0.01022576,
      "balance_loss_clip": 1.05434394,
      "balance_loss_mlp": 1.01521778,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.7653930695851843,
      "language_loss": 0.79002702,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81204647,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.6672730445861816
    },
    {
      "auxiliary_loss_clip": 0.0117985,
      "auxiliary_loss_mlp": 0.01123946,
      "balance_loss_clip": 1.01307237,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 1.8957152432510225,
      "language_loss": 0.75915694,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.78219491,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 3.6897222995758057
    },
    {
      "auxiliary_loss_clip": 0.01168387,
      "auxiliary_loss_mlp": 0.01023169,
      "balance_loss_clip": 0.97287267,
      "balance_loss_mlp": 1.01616502,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.711262792152538,
      "language_loss": 0.70192617,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.72384173,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 2.7246758937835693
    },
    {
      "auxiliary_loss_clip": 0.01178909,
      "auxiliary_loss_mlp": 0.01024482,
      "balance_loss_clip": 1.05121279,
      "balance_loss_mlp": 1.01646531,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 2.2408592939358947,
      "language_loss": 0.77749336,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.79952729,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.611072540283203
    },
    {
      "auxiliary_loss_clip": 0.01171348,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 0.97547257,
      "balance_loss_mlp": 1.01959443,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 2.6884794781548638,
      "language_loss": 0.71287394,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73485661,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 2.737849473953247
    },
    {
      "auxiliary_loss_clip": 0.01177404,
      "auxiliary_loss_mlp": 0.01026349,
      "balance_loss_clip": 1.05235338,
      "balance_loss_mlp": 1.01885629,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 4.349173460847881,
      "language_loss": 0.77979296,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.80183053,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 3.515509843826294
    },
    {
      "auxiliary_loss_clip": 0.01176535,
      "auxiliary_loss_mlp": 0.01028574,
      "balance_loss_clip": 1.05319333,
      "balance_loss_mlp": 1.02027738,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 1.6182322862727434,
      "language_loss": 0.70642781,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.72847891,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 2.629213333129883
    },
    {
      "auxiliary_loss_clip": 0.01176092,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 1.01563585,
      "balance_loss_mlp": 1.02183855,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 2.1520726234473173,
      "language_loss": 0.72199297,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74404585,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.6958205699920654
    },
    {
      "auxiliary_loss_clip": 0.01169514,
      "auxiliary_loss_mlp": 0.01027376,
      "balance_loss_clip": 1.01380754,
      "balance_loss_mlp": 1.0204891,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.768618946479683,
      "language_loss": 0.88023567,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90220451,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.6087028980255127
    },
    {
      "auxiliary_loss_clip": 0.01175039,
      "auxiliary_loss_mlp": 0.01024836,
      "balance_loss_clip": 0.97568166,
      "balance_loss_mlp": 1.01759076,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 1.9245039294541366,
      "language_loss": 0.68946528,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.71146405,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.739171028137207
    },
    {
      "auxiliary_loss_clip": 0.01171699,
      "auxiliary_loss_mlp": 0.01122967,
      "balance_loss_clip": 1.0102824,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 1.696850602643885,
      "language_loss": 0.79443377,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.81738043,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.692054271697998
    },
    {
      "auxiliary_loss_clip": 0.01171754,
      "auxiliary_loss_mlp": 0.01122943,
      "balance_loss_clip": 1.01415896,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 1.7175384019862086,
      "language_loss": 0.72592366,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.74887061,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 2.660144329071045
    },
    {
      "auxiliary_loss_clip": 0.01174074,
      "auxiliary_loss_mlp": 0.01030215,
      "balance_loss_clip": 1.01397979,
      "balance_loss_mlp": 1.02245748,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 1.7814429985361522,
      "language_loss": 0.83333892,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.85538179,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 2.6728224754333496
    },
    {
      "auxiliary_loss_clip": 0.01164183,
      "auxiliary_loss_mlp": 0.01030915,
      "balance_loss_clip": 0.85438335,
      "balance_loss_mlp": 1.02245736,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 1.80580381205194,
      "language_loss": 0.78302014,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.8049711,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.924346923828125
    },
    {
      "auxiliary_loss_clip": 0.01175276,
      "auxiliary_loss_mlp": 0.01030112,
      "balance_loss_clip": 0.97833598,
      "balance_loss_mlp": 1.02175534,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 6.684273886854852,
      "language_loss": 0.86793184,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.88998568,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.6934430599212646
    },
    {
      "auxiliary_loss_clip": 0.01176244,
      "auxiliary_loss_mlp": 0.01029742,
      "balance_loss_clip": 1.01333332,
      "balance_loss_mlp": 1.02282155,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.4495018435377771,
      "language_loss": 0.76195264,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.78401256,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 2.6912269592285156
    },
    {
      "auxiliary_loss_clip": 0.01160488,
      "auxiliary_loss_mlp": 0.01023869,
      "balance_loss_clip": 0.97069275,
      "balance_loss_mlp": 1.01609087,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 1.9657680457449707,
      "language_loss": 0.68487048,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.70671403,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 2.72220778465271
    },
    {
      "auxiliary_loss_clip": 0.01165764,
      "auxiliary_loss_mlp": 0.01033075,
      "balance_loss_clip": 0.89322937,
      "balance_loss_mlp": 1.0254935,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 1.966968897170057,
      "language_loss": 0.80005044,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.82203883,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 2.7411866188049316
    },
    {
      "auxiliary_loss_clip": 0.01169332,
      "auxiliary_loss_mlp": 0.0102784,
      "balance_loss_clip": 0.97041273,
      "balance_loss_mlp": 1.02044892,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 1.6115406690379366,
      "language_loss": 0.66332734,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.6852991,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 2.812331438064575
    },
    {
      "auxiliary_loss_clip": 0.01177178,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 0.93666941,
      "balance_loss_mlp": 1.0232079,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 1.9869339843036018,
      "language_loss": 0.84855676,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87063587,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 2.656986713409424
    },
    {
      "auxiliary_loss_clip": 0.0116519,
      "auxiliary_loss_mlp": 0.01028144,
      "balance_loss_clip": 0.9344722,
      "balance_loss_mlp": 1.02040398,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 1.8482775140224128,
      "language_loss": 0.89574754,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91768086,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.797795295715332
    },
    {
      "auxiliary_loss_clip": 0.01176627,
      "auxiliary_loss_mlp": 0.0112268,
      "balance_loss_clip": 0.9769243,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 1.7333984440565384,
      "language_loss": 0.75128341,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.77427649,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 3.6578288078308105
    },
    {
      "auxiliary_loss_clip": 0.01178369,
      "auxiliary_loss_mlp": 0.01035504,
      "balance_loss_clip": 0.97626805,
      "balance_loss_mlp": 1.02812505,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 1.6557563097158754,
      "language_loss": 0.75568533,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.77782404,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.6507580280303955
    },
    {
      "auxiliary_loss_clip": 0.01166311,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 0.85926592,
      "balance_loss_mlp": 1.02032328,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.6001740020390396,
      "language_loss": 0.76787865,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.78982204,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 3.825134754180908
    },
    {
      "auxiliary_loss_clip": 0.01180363,
      "auxiliary_loss_mlp": 0.01030278,
      "balance_loss_clip": 1.01388633,
      "balance_loss_mlp": 1.02316153,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 1.8077171998963375,
      "language_loss": 0.77356327,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.79566973,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.645902395248413
    },
    {
      "auxiliary_loss_clip": 0.01176641,
      "auxiliary_loss_mlp": 0.01022029,
      "balance_loss_clip": 0.97296607,
      "balance_loss_mlp": 1.0150013,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.6842932357843434,
      "language_loss": 0.76180017,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78378683,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.7064061164855957
    },
    {
      "auxiliary_loss_clip": 0.01175955,
      "auxiliary_loss_mlp": 0.01026544,
      "balance_loss_clip": 1.04979765,
      "balance_loss_mlp": 1.01851487,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 2.162753702113439,
      "language_loss": 0.80520213,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.82722712,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 3.667018413543701
    },
    {
      "auxiliary_loss_clip": 0.0117162,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.04925799,
      "balance_loss_mlp": 1.02330577,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 2.5508339104383504,
      "language_loss": 0.65884477,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.68087196,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.7294063568115234
    },
    {
      "auxiliary_loss_clip": 0.01168065,
      "auxiliary_loss_mlp": 0.01025442,
      "balance_loss_clip": 0.97604877,
      "balance_loss_mlp": 1.01816452,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 1.834788663508302,
      "language_loss": 0.80685198,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82878709,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.6451730728149414
    },
    {
      "auxiliary_loss_clip": 0.01181024,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 0.97470766,
      "balance_loss_mlp": 1.01651347,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 1.9877433010908132,
      "language_loss": 0.78455853,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.80660987,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 2.6134426593780518
    },
    {
      "auxiliary_loss_clip": 0.01170514,
      "auxiliary_loss_mlp": 0.01029967,
      "balance_loss_clip": 0.89408457,
      "balance_loss_mlp": 1.02222395,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 1.8384308692238116,
      "language_loss": 0.83439642,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.85640121,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.771425485610962
    },
    {
      "auxiliary_loss_clip": 0.01176482,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.05209649,
      "balance_loss_mlp": 1.01822984,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 1.9782442116962915,
      "language_loss": 0.8159368,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.83795285,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 3.5453426837921143
    },
    {
      "auxiliary_loss_clip": 0.01155828,
      "auxiliary_loss_mlp": 0.01030293,
      "balance_loss_clip": 0.74194151,
      "balance_loss_mlp": 1.02318239,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.041498967250096,
      "language_loss": 0.82878661,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.85064781,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 3.050366163253784
    },
    {
      "auxiliary_loss_clip": 0.01162248,
      "auxiliary_loss_mlp": 0.01024685,
      "balance_loss_clip": 0.85933834,
      "balance_loss_mlp": 1.01749969,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 1.5510041061245226,
      "language_loss": 0.78399694,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.8058663,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 2.969479560852051
    },
    {
      "auxiliary_loss_clip": 0.01177216,
      "auxiliary_loss_mlp": 0.0102937,
      "balance_loss_clip": 1.05080581,
      "balance_loss_mlp": 1.02194965,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 2.2406795239731965,
      "language_loss": 0.86131263,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.88337851,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.578857421875
    },
    {
      "auxiliary_loss_clip": 0.01178699,
      "auxiliary_loss_mlp": 0.0112257,
      "balance_loss_clip": 1.01432681,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 1.7436190610408238,
      "language_loss": 0.79974151,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.8227542,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 2.712008476257324
    },
    {
      "auxiliary_loss_clip": 0.01175056,
      "auxiliary_loss_mlp": 0.01025852,
      "balance_loss_clip": 0.9704538,
      "balance_loss_mlp": 1.01881588,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 2.123773620887498,
      "language_loss": 0.72942007,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75142908,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.6839191913604736
    },
    {
      "auxiliary_loss_clip": 0.01175568,
      "auxiliary_loss_mlp": 0.01028522,
      "balance_loss_clip": 0.97428876,
      "balance_loss_mlp": 1.02092266,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 1.63766792420767,
      "language_loss": 0.77114534,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79318619,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 3.426009178161621
    },
    {
      "auxiliary_loss_clip": 0.01173684,
      "auxiliary_loss_mlp": 0.01025364,
      "balance_loss_clip": 1.05126226,
      "balance_loss_mlp": 1.01844084,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 1.8999275287796757,
      "language_loss": 0.82356691,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84555733,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 2.634549617767334
    },
    {
      "auxiliary_loss_clip": 0.01171795,
      "auxiliary_loss_mlp": 0.01122768,
      "balance_loss_clip": 0.93294072,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 2.153038445079365,
      "language_loss": 0.78286183,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.80580741,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.733930826187134
    },
    {
      "auxiliary_loss_clip": 0.01176208,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 1.05164039,
      "balance_loss_mlp": 1.02096939,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 1.703110592673255,
      "language_loss": 0.85320079,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.87524438,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.66784405708313
    },
    {
      "auxiliary_loss_clip": 0.01179404,
      "auxiliary_loss_mlp": 0.0102966,
      "balance_loss_clip": 0.97579992,
      "balance_loss_mlp": 1.02176237,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 2.1697795204871864,
      "language_loss": 0.81677961,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.83887017,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 2.616541624069214
    },
    {
      "auxiliary_loss_clip": 0.01080059,
      "auxiliary_loss_mlp": 0.01005153,
      "balance_loss_clip": 1.02181435,
      "balance_loss_mlp": 1.00346017,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.7265061635175725,
      "language_loss": 0.63223684,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65308893,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 3.325915813446045
    },
    {
      "auxiliary_loss_clip": 0.01178127,
      "auxiliary_loss_mlp": 0.01026521,
      "balance_loss_clip": 0.97758466,
      "balance_loss_mlp": 1.01898742,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 1.6719191684765227,
      "language_loss": 0.76723278,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.78927934,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 2.726456880569458
    },
    {
      "auxiliary_loss_clip": 0.01171369,
      "auxiliary_loss_mlp": 0.01029249,
      "balance_loss_clip": 0.97115201,
      "balance_loss_mlp": 1.02179205,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.555448533144301,
      "language_loss": 0.75478524,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.77679145,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 2.6975786685943604
    },
    {
      "auxiliary_loss_clip": 0.01176551,
      "auxiliary_loss_mlp": 0.01027785,
      "balance_loss_clip": 0.97659367,
      "balance_loss_mlp": 1.02090979,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 1.6283907473560677,
      "language_loss": 0.84424567,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.86628902,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.68530011177063
    },
    {
      "auxiliary_loss_clip": 0.01169874,
      "auxiliary_loss_mlp": 0.01028035,
      "balance_loss_clip": 0.97347248,
      "balance_loss_mlp": 1.02029228,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.605939172476789,
      "language_loss": 0.7906847,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.81266379,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 3.6937475204467773
    },
    {
      "auxiliary_loss_clip": 0.01166822,
      "auxiliary_loss_mlp": 0.01026525,
      "balance_loss_clip": 0.93525976,
      "balance_loss_mlp": 1.01948524,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 1.8168645193206585,
      "language_loss": 0.83937877,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86131221,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 2.7487196922302246
    },
    {
      "auxiliary_loss_clip": 0.01078887,
      "auxiliary_loss_mlp": 0.0100425,
      "balance_loss_clip": 1.02083111,
      "balance_loss_mlp": 1.002581,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.772296205842347,
      "language_loss": 0.57891464,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59974599,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 3.1754908561706543
    },
    {
      "auxiliary_loss_clip": 0.01179295,
      "auxiliary_loss_mlp": 0.0102427,
      "balance_loss_clip": 1.01316631,
      "balance_loss_mlp": 1.01724863,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 1.7736352180502037,
      "language_loss": 0.72689593,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.74893159,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 3.919572114944458
    },
    {
      "auxiliary_loss_clip": 0.01170532,
      "auxiliary_loss_mlp": 0.010277,
      "balance_loss_clip": 1.01369524,
      "balance_loss_mlp": 1.0199455,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 1.5140074148186125,
      "language_loss": 0.72320449,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74518687,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.8116278648376465
    },
    {
      "auxiliary_loss_clip": 0.01176362,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.05139709,
      "balance_loss_mlp": 1.01994109,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 1.8988315998827008,
      "language_loss": 0.78051829,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80255878,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.667147397994995
    },
    {
      "auxiliary_loss_clip": 0.01177123,
      "auxiliary_loss_mlp": 0.01032982,
      "balance_loss_clip": 1.01479435,
      "balance_loss_mlp": 1.02532828,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 2.0991554573876767,
      "language_loss": 0.74023288,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76233393,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 3.643336057662964
    },
    {
      "auxiliary_loss_clip": 0.01176196,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.05180657,
      "balance_loss_mlp": 1.02039814,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.5502106069908697,
      "language_loss": 0.79421103,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.81624752,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.600972890853882
    },
    {
      "auxiliary_loss_clip": 0.01178785,
      "auxiliary_loss_mlp": 0.01027051,
      "balance_loss_clip": 1.05240667,
      "balance_loss_mlp": 1.0191946,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 2.008552739490441,
      "language_loss": 0.72706461,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.74912292,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.606633424758911
    },
    {
      "auxiliary_loss_clip": 0.01173313,
      "auxiliary_loss_mlp": 0.01024672,
      "balance_loss_clip": 1.01474214,
      "balance_loss_mlp": 1.01672101,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 1.9312604726171245,
      "language_loss": 0.88236755,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90434742,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 2.6548893451690674
    },
    {
      "auxiliary_loss_clip": 0.01174472,
      "auxiliary_loss_mlp": 0.01029917,
      "balance_loss_clip": 1.05059087,
      "balance_loss_mlp": 1.02275872,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 1.75908595286192,
      "language_loss": 0.70416701,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72621095,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 2.64452862739563
    },
    {
      "auxiliary_loss_clip": 0.01157591,
      "auxiliary_loss_mlp": 0.01026434,
      "balance_loss_clip": 0.93031359,
      "balance_loss_mlp": 1.01887929,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 1.4690718814564754,
      "language_loss": 0.73710543,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.7589457,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 3.6835551261901855
    },
    {
      "auxiliary_loss_clip": 0.0117639,
      "auxiliary_loss_mlp": 0.0112331,
      "balance_loss_clip": 0.97593856,
      "balance_loss_mlp": 0.0,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 2.6870365865263204,
      "language_loss": 0.78785241,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.81084943,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.7090983390808105
    },
    {
      "auxiliary_loss_clip": 0.01174366,
      "auxiliary_loss_mlp": 0.01031338,
      "balance_loss_clip": 1.01148331,
      "balance_loss_mlp": 1.02349997,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.636380960497627,
      "language_loss": 0.7903772,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.81243426,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.644164562225342
    },
    {
      "auxiliary_loss_clip": 0.01174692,
      "auxiliary_loss_mlp": 0.01123527,
      "balance_loss_clip": 1.01196289,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 2.095677284868044,
      "language_loss": 0.74949086,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.7724731,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 2.6406326293945312
    },
    {
      "auxiliary_loss_clip": 0.01174549,
      "auxiliary_loss_mlp": 0.01122807,
      "balance_loss_clip": 1.01448631,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.3755020765127866,
      "language_loss": 0.76536191,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.7883355,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.902128219604492
    },
    {
      "auxiliary_loss_clip": 0.01167752,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 0.89501137,
      "balance_loss_mlp": 1.01870465,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 1.6873951287372453,
      "language_loss": 0.75475812,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77670002,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 2.753251791000366
    },
    {
      "auxiliary_loss_clip": 0.01170325,
      "auxiliary_loss_mlp": 0.01030813,
      "balance_loss_clip": 0.93354046,
      "balance_loss_mlp": 1.02202129,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 1.8258056897101305,
      "language_loss": 0.78151876,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.8035301,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.6966466903686523
    },
    {
      "auxiliary_loss_clip": 0.01176585,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.01372099,
      "balance_loss_mlp": 1.01986194,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.3938555755131437,
      "language_loss": 0.77276731,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.79480457,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.657944440841675
    },
    {
      "auxiliary_loss_clip": 0.01161483,
      "auxiliary_loss_mlp": 0.0102572,
      "balance_loss_clip": 0.97118139,
      "balance_loss_mlp": 1.01841271,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 1.916159954012353,
      "language_loss": 0.68827564,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.71014774,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 2.702550172805786
    },
    {
      "auxiliary_loss_clip": 0.01156365,
      "auxiliary_loss_mlp": 0.01122968,
      "balance_loss_clip": 0.85882831,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 2.262030781485172,
      "language_loss": 0.76221383,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.78500712,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.7546727657318115
    },
    {
      "auxiliary_loss_clip": 0.01179234,
      "auxiliary_loss_mlp": 0.01031356,
      "balance_loss_clip": 1.05206978,
      "balance_loss_mlp": 1.02305961,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 2.3021120063559413,
      "language_loss": 0.84553158,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86763746,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.609367847442627
    },
    {
      "auxiliary_loss_clip": 0.01171959,
      "auxiliary_loss_mlp": 0.01028624,
      "balance_loss_clip": 1.01387024,
      "balance_loss_mlp": 1.02157271,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 2.2210536598912465,
      "language_loss": 0.77045023,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79245603,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 2.608546018600464
    },
    {
      "auxiliary_loss_clip": 0.01163145,
      "auxiliary_loss_mlp": 0.01025182,
      "balance_loss_clip": 0.97063953,
      "balance_loss_mlp": 1.01767528,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 2.187950531908411,
      "language_loss": 0.64064765,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.6625309,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 2.6406452655792236
    },
    {
      "auxiliary_loss_clip": 0.0117788,
      "auxiliary_loss_mlp": 0.01024186,
      "balance_loss_clip": 0.975389,
      "balance_loss_mlp": 1.01661611,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 1.6324575848669702,
      "language_loss": 0.75272155,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77474225,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 2.7243690490722656
    },
    {
      "auxiliary_loss_clip": 0.01176324,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.05154669,
      "balance_loss_mlp": 1.018062,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 1.7612610542800171,
      "language_loss": 0.80545497,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.82747245,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.657492160797119
    },
    {
      "auxiliary_loss_clip": 0.01172206,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 0.93352902,
      "balance_loss_mlp": 1.0181191,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 2.1015356699156875,
      "language_loss": 0.81025708,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.83224553,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 2.680356502532959
    },
    {
      "auxiliary_loss_clip": 0.01175487,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 0.97625685,
      "balance_loss_mlp": 1.01843786,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.6824986905686878,
      "language_loss": 0.7886498,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.81066597,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 3.9915363788604736
    },
    {
      "auxiliary_loss_clip": 0.01171333,
      "auxiliary_loss_mlp": 0.01026961,
      "balance_loss_clip": 0.93472207,
      "balance_loss_mlp": 1.01885152,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 2.098477128763159,
      "language_loss": 0.73724341,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.75922638,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.654310941696167
    },
    {
      "auxiliary_loss_clip": 0.01089691,
      "auxiliary_loss_mlp": 0.01004276,
      "balance_loss_clip": 0.88324785,
      "balance_loss_mlp": 1.00244009,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7332107769124592,
      "language_loss": 0.61671722,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63765693,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 4.1942760944366455
    },
    {
      "auxiliary_loss_clip": 0.01152477,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 0.92859226,
      "balance_loss_mlp": 1.02305436,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 1.9985790150830982,
      "language_loss": 0.60854304,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.63037646,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.71940279006958
    },
    {
      "auxiliary_loss_clip": 0.01174261,
      "auxiliary_loss_mlp": 0.01028811,
      "balance_loss_clip": 1.04930758,
      "balance_loss_mlp": 1.02126837,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 2.064893399502752,
      "language_loss": 0.817873,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.83990377,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.6464736461639404
    },
    {
      "auxiliary_loss_clip": 0.01174159,
      "auxiliary_loss_mlp": 0.01023177,
      "balance_loss_clip": 1.01226902,
      "balance_loss_mlp": 1.01573813,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 1.6722189764060558,
      "language_loss": 0.806714,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.82868737,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 3.5595107078552246
    },
    {
      "auxiliary_loss_clip": 0.01173591,
      "auxiliary_loss_mlp": 0.01030127,
      "balance_loss_clip": 1.0146035,
      "balance_loss_mlp": 1.02323651,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 1.5186308869019698,
      "language_loss": 0.84390235,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.86593956,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.936065912246704
    },
    {
      "auxiliary_loss_clip": 0.01176734,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 0.97144133,
      "balance_loss_mlp": 1.02136946,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 1.9812520305336985,
      "language_loss": 0.69923574,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.72129416,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.736241102218628
    },
    {
      "auxiliary_loss_clip": 0.01172704,
      "auxiliary_loss_mlp": 0.01027731,
      "balance_loss_clip": 0.9414916,
      "balance_loss_mlp": 1.02052224,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 1.5535249077124589,
      "language_loss": 0.73960257,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76160693,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 2.686276435852051
    },
    {
      "auxiliary_loss_clip": 0.01175296,
      "auxiliary_loss_mlp": 0.0102008,
      "balance_loss_clip": 1.05104613,
      "balance_loss_mlp": 1.01302266,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 1.5842096598008277,
      "language_loss": 0.70305514,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.7250089,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.6166276931762695
    },
    {
      "auxiliary_loss_clip": 0.01174355,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 0.93532497,
      "balance_loss_mlp": 1.02056026,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 1.808944597152072,
      "language_loss": 0.84909129,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.8711133,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 3.647942066192627
    },
    {
      "auxiliary_loss_clip": 0.01170224,
      "auxiliary_loss_mlp": 0.0103345,
      "balance_loss_clip": 0.97096682,
      "balance_loss_mlp": 1.02541482,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.3242568355559454,
      "language_loss": 0.70832324,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.73035997,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.702502489089966
    },
    {
      "auxiliary_loss_clip": 0.01171661,
      "auxiliary_loss_mlp": 0.01029573,
      "balance_loss_clip": 1.01367724,
      "balance_loss_mlp": 1.02169323,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 1.7258786586500496,
      "language_loss": 0.74136889,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.76338124,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 2.6738874912261963
    },
    {
      "auxiliary_loss_clip": 0.01082221,
      "auxiliary_loss_mlp": 0.01003985,
      "balance_loss_clip": 0.90636522,
      "balance_loss_mlp": 1.00210178,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.7012205640638194,
      "language_loss": 0.52395421,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54481632,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.2287280559539795
    },
    {
      "auxiliary_loss_clip": 0.01089328,
      "auxiliary_loss_mlp": 0.01001876,
      "balance_loss_clip": 0.86923862,
      "balance_loss_mlp": 0.99995708,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.7123221283698115,
      "language_loss": 0.54504371,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56595576,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.351167917251587
    },
    {
      "auxiliary_loss_clip": 0.01179354,
      "auxiliary_loss_mlp": 0.01027154,
      "balance_loss_clip": 1.01503849,
      "balance_loss_mlp": 1.01949799,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.8829361244574536,
      "language_loss": 0.74914551,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.77121067,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.6598165035247803
    },
    {
      "auxiliary_loss_clip": 0.01168936,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 0.97148073,
      "balance_loss_mlp": 1.01911378,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 1.9746558831261456,
      "language_loss": 0.84356582,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86551869,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 2.7135417461395264
    },
    {
      "auxiliary_loss_clip": 0.01174725,
      "auxiliary_loss_mlp": 0.01030931,
      "balance_loss_clip": 1.01349711,
      "balance_loss_mlp": 1.023278,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 2.040787505857809,
      "language_loss": 0.82815003,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.85020655,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.6642723083496094
    },
    {
      "auxiliary_loss_clip": 0.01169037,
      "auxiliary_loss_mlp": 0.01027066,
      "balance_loss_clip": 0.93497324,
      "balance_loss_mlp": 1.01892376,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 2.502468809962186,
      "language_loss": 0.83213311,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85409415,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.8164279460906982
    },
    {
      "auxiliary_loss_clip": 0.0116814,
      "auxiliary_loss_mlp": 0.01026636,
      "balance_loss_clip": 0.97389364,
      "balance_loss_mlp": 1.01886964,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 2.239313872592076,
      "language_loss": 0.77176821,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79371601,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 2.659482002258301
    },
    {
      "auxiliary_loss_clip": 0.01179858,
      "auxiliary_loss_mlp": 0.01034233,
      "balance_loss_clip": 0.93499875,
      "balance_loss_mlp": 1.02660346,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 1.6531155045763637,
      "language_loss": 0.70406425,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.72620517,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.799955129623413
    },
    {
      "auxiliary_loss_clip": 0.011702,
      "auxiliary_loss_mlp": 0.01027158,
      "balance_loss_clip": 0.97227758,
      "balance_loss_mlp": 1.01918852,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 1.7091106536903062,
      "language_loss": 0.77525175,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.79722536,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 2.6622867584228516
    },
    {
      "auxiliary_loss_clip": 0.0117241,
      "auxiliary_loss_mlp": 0.0102992,
      "balance_loss_clip": 1.01198387,
      "balance_loss_mlp": 1.02280366,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 1.9982203957518048,
      "language_loss": 0.7732408,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.79526412,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 2.6199238300323486
    },
    {
      "auxiliary_loss_clip": 0.01180904,
      "auxiliary_loss_mlp": 0.01030747,
      "balance_loss_clip": 0.97584403,
      "balance_loss_mlp": 1.02405024,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 1.5449545349198728,
      "language_loss": 0.84951645,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87163299,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 2.637514591217041
    },
    {
      "auxiliary_loss_clip": 0.01168331,
      "auxiliary_loss_mlp": 0.0102828,
      "balance_loss_clip": 0.97367656,
      "balance_loss_mlp": 1.02114248,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 2.553458182709257,
      "language_loss": 0.72286755,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.74483365,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 2.763655662536621
    },
    {
      "auxiliary_loss_clip": 0.01172908,
      "auxiliary_loss_mlp": 0.01025018,
      "balance_loss_clip": 0.93540937,
      "balance_loss_mlp": 1.01750779,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 2.055251425670415,
      "language_loss": 0.84411085,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.86609012,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.820150852203369
    },
    {
      "auxiliary_loss_clip": 0.01173914,
      "auxiliary_loss_mlp": 0.01021639,
      "balance_loss_clip": 1.01071763,
      "balance_loss_mlp": 1.01479948,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 4.296330765720839,
      "language_loss": 0.77471471,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.7966702,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 3.535870313644409
    },
    {
      "auxiliary_loss_clip": 0.01173622,
      "auxiliary_loss_mlp": 0.0102392,
      "balance_loss_clip": 1.01279688,
      "balance_loss_mlp": 1.01633191,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.637935263666629,
      "language_loss": 0.76087087,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.78284633,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.671049118041992
    },
    {
      "auxiliary_loss_clip": 0.01080173,
      "auxiliary_loss_mlp": 0.01001319,
      "balance_loss_clip": 0.86714566,
      "balance_loss_mlp": 0.99964964,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.8320131636954873,
      "language_loss": 0.57913792,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.59995282,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.353775978088379
    },
    {
      "auxiliary_loss_clip": 0.01174688,
      "auxiliary_loss_mlp": 0.01023458,
      "balance_loss_clip": 1.00954878,
      "balance_loss_mlp": 1.01593041,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 1.845718073563209,
      "language_loss": 0.78798044,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.80996192,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 3.7518651485443115
    },
    {
      "auxiliary_loss_clip": 0.01172506,
      "auxiliary_loss_mlp": 0.01025369,
      "balance_loss_clip": 1.01049793,
      "balance_loss_mlp": 1.01762605,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 3.0045373372968442,
      "language_loss": 0.73972118,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76169991,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.680792808532715
    },
    {
      "auxiliary_loss_clip": 0.01177341,
      "auxiliary_loss_mlp": 0.01022902,
      "balance_loss_clip": 1.01385355,
      "balance_loss_mlp": 1.01555908,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 1.872486206365953,
      "language_loss": 0.75450456,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77650696,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 3.6994097232818604
    },
    {
      "auxiliary_loss_clip": 0.01158868,
      "auxiliary_loss_mlp": 0.01022673,
      "balance_loss_clip": 0.93328178,
      "balance_loss_mlp": 1.01551175,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 1.7986180851475766,
      "language_loss": 0.69309908,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71491444,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.6972944736480713
    },
    {
      "auxiliary_loss_clip": 0.01174133,
      "auxiliary_loss_mlp": 0.01123087,
      "balance_loss_clip": 0.89418137,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 1.7052845526763487,
      "language_loss": 0.65013182,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.67310393,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.8134331703186035
    },
    {
      "auxiliary_loss_clip": 0.01164705,
      "auxiliary_loss_mlp": 0.01025421,
      "balance_loss_clip": 0.97110009,
      "balance_loss_mlp": 1.01835489,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.1108560020802623,
      "language_loss": 0.79798776,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.81988895,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.7388062477111816
    },
    {
      "auxiliary_loss_clip": 0.01087043,
      "auxiliary_loss_mlp": 0.01001387,
      "balance_loss_clip": 0.8663038,
      "balance_loss_mlp": 0.99947929,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.9374898142105726,
      "language_loss": 0.61982477,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.64070904,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 3.1736836433410645
    },
    {
      "auxiliary_loss_clip": 0.01162013,
      "auxiliary_loss_mlp": 0.01025319,
      "balance_loss_clip": 0.97041166,
      "balance_loss_mlp": 1.01810145,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.634851722048585,
      "language_loss": 0.7875486,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.80942196,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 3.6234967708587646
    },
    {
      "auxiliary_loss_clip": 0.01168249,
      "auxiliary_loss_mlp": 0.01029414,
      "balance_loss_clip": 0.97577298,
      "balance_loss_mlp": 1.02200484,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 1.8485848912632932,
      "language_loss": 0.74900115,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.77097774,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 2.835470676422119
    },
    {
      "auxiliary_loss_clip": 0.0117566,
      "auxiliary_loss_mlp": 0.01026706,
      "balance_loss_clip": 1.04983056,
      "balance_loss_mlp": 1.01896334,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 1.8176231855121192,
      "language_loss": 0.72664905,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.74867272,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.628831148147583
    },
    {
      "auxiliary_loss_clip": 0.01172357,
      "auxiliary_loss_mlp": 0.01122697,
      "balance_loss_clip": 1.05023599,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 1.7939366361463975,
      "language_loss": 0.7403065,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.76325703,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 2.7795724868774414
    },
    {
      "auxiliary_loss_clip": 0.01172058,
      "auxiliary_loss_mlp": 0.01023489,
      "balance_loss_clip": 0.97360104,
      "balance_loss_mlp": 1.01628244,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 2.457034564770766,
      "language_loss": 0.75447822,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.77643371,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 2.755877733230591
    },
    {
      "auxiliary_loss_clip": 0.01171583,
      "auxiliary_loss_mlp": 0.0102319,
      "balance_loss_clip": 0.97232866,
      "balance_loss_mlp": 1.01593614,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 2.395900295951043,
      "language_loss": 0.76601112,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78795886,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.6745994091033936
    },
    {
      "auxiliary_loss_clip": 0.01178779,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.01188254,
      "balance_loss_mlp": 1.0223918,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 1.760400458732099,
      "language_loss": 0.78603333,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.80811965,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 2.692812919616699
    },
    {
      "auxiliary_loss_clip": 0.01173977,
      "auxiliary_loss_mlp": 0.01024663,
      "balance_loss_clip": 1.05009508,
      "balance_loss_mlp": 1.01758194,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 1.7886536092631484,
      "language_loss": 0.70725036,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.72923672,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.691563129425049
    },
    {
      "auxiliary_loss_clip": 0.01166154,
      "auxiliary_loss_mlp": 0.01023771,
      "balance_loss_clip": 0.9735747,
      "balance_loss_mlp": 1.01707792,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 1.741458799517779,
      "language_loss": 0.78946877,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81136805,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.6733522415161133
    },
    {
      "auxiliary_loss_clip": 0.01165563,
      "auxiliary_loss_mlp": 0.01030126,
      "balance_loss_clip": 0.89686561,
      "balance_loss_mlp": 1.02303648,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 1.6167702146003675,
      "language_loss": 0.80601996,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.82797682,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 2.7518253326416016
    },
    {
      "auxiliary_loss_clip": 0.01168205,
      "auxiliary_loss_mlp": 0.01025251,
      "balance_loss_clip": 0.97325885,
      "balance_loss_mlp": 1.01803327,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 1.598633741378701,
      "language_loss": 0.79528737,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81722194,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 2.7287440299987793
    },
    {
      "auxiliary_loss_clip": 0.01171144,
      "auxiliary_loss_mlp": 0.01028065,
      "balance_loss_clip": 0.9752847,
      "balance_loss_mlp": 1.02113342,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 1.7605210563620335,
      "language_loss": 0.72687989,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.74887198,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.7073373794555664
    },
    {
      "auxiliary_loss_clip": 0.01158318,
      "auxiliary_loss_mlp": 0.01022617,
      "balance_loss_clip": 0.93279433,
      "balance_loss_mlp": 1.0157057,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 2.127964023921297,
      "language_loss": 0.71547985,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.73728925,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 2.803891897201538
    },
    {
      "auxiliary_loss_clip": 0.0117032,
      "auxiliary_loss_mlp": 0.01028901,
      "balance_loss_clip": 0.89650857,
      "balance_loss_mlp": 1.02112842,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.8131748884629242,
      "language_loss": 0.60865182,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.63064402,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 2.8246636390686035
    },
    {
      "auxiliary_loss_clip": 0.01174365,
      "auxiliary_loss_mlp": 0.01024093,
      "balance_loss_clip": 1.05088782,
      "balance_loss_mlp": 1.01611781,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 1.8230174396850982,
      "language_loss": 0.72100914,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.74299377,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.647641658782959
    },
    {
      "auxiliary_loss_clip": 0.01162476,
      "auxiliary_loss_mlp": 0.01032869,
      "balance_loss_clip": 1.01133835,
      "balance_loss_mlp": 1.02581799,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 1.7507267275533096,
      "language_loss": 0.78236628,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80431974,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.6495509147644043
    },
    {
      "auxiliary_loss_clip": 0.01167442,
      "auxiliary_loss_mlp": 0.01020367,
      "balance_loss_clip": 0.93297613,
      "balance_loss_mlp": 1.01375437,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 1.5623884202614984,
      "language_loss": 0.67471159,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.69658971,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 3.8524680137634277
    },
    {
      "auxiliary_loss_clip": 0.01168623,
      "auxiliary_loss_mlp": 0.01029879,
      "balance_loss_clip": 1.01091933,
      "balance_loss_mlp": 1.02200532,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.7105294141859009,
      "language_loss": 0.71341699,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.73540205,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.631053924560547
    },
    {
      "auxiliary_loss_clip": 0.01174555,
      "auxiliary_loss_mlp": 0.01029616,
      "balance_loss_clip": 0.89707023,
      "balance_loss_mlp": 1.0216645,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 1.825955303961391,
      "language_loss": 0.76442218,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.78646386,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.7532148361206055
    },
    {
      "auxiliary_loss_clip": 0.01175113,
      "auxiliary_loss_mlp": 0.01024395,
      "balance_loss_clip": 1.05115891,
      "balance_loss_mlp": 1.01710582,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.8798024706986955,
      "language_loss": 0.73285168,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75484681,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 3.6164915561676025
    },
    {
      "auxiliary_loss_clip": 0.01167339,
      "auxiliary_loss_mlp": 0.01021561,
      "balance_loss_clip": 0.97003496,
      "balance_loss_mlp": 1.01425695,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 3.014713508157937,
      "language_loss": 0.74454135,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76643038,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.6845834255218506
    },
    {
      "auxiliary_loss_clip": 0.01160566,
      "auxiliary_loss_mlp": 0.01022411,
      "balance_loss_clip": 0.97152942,
      "balance_loss_mlp": 1.01566112,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.0534662855679393,
      "language_loss": 0.8321383,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.85396808,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 3.693748950958252
    },
    {
      "auxiliary_loss_clip": 0.01165625,
      "auxiliary_loss_mlp": 0.01030776,
      "balance_loss_clip": 0.97259009,
      "balance_loss_mlp": 1.02336168,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.4032260541081658,
      "language_loss": 0.73932952,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76129353,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.7303383350372314
    },
    {
      "auxiliary_loss_clip": 0.01176051,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.05142665,
      "balance_loss_mlp": 1.01750326,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 1.640759825219395,
      "language_loss": 0.71583086,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.7378366,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.64003586769104
    },
    {
      "auxiliary_loss_clip": 0.01173797,
      "auxiliary_loss_mlp": 0.01029163,
      "balance_loss_clip": 0.9766683,
      "balance_loss_mlp": 1.02148604,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 1.7007261529046787,
      "language_loss": 0.74511206,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.7671417,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 2.8421645164489746
    },
    {
      "auxiliary_loss_clip": 0.01177356,
      "auxiliary_loss_mlp": 0.01024999,
      "balance_loss_clip": 1.0107646,
      "balance_loss_mlp": 1.01741147,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 23.097462559625914,
      "language_loss": 0.76442361,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.78644717,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.750303030014038
    },
    {
      "auxiliary_loss_clip": 0.01168721,
      "auxiliary_loss_mlp": 0.0102605,
      "balance_loss_clip": 0.9335447,
      "balance_loss_mlp": 1.01822352,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.0846820517224653,
      "language_loss": 0.68424958,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70619732,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 3.561199903488159
    },
    {
      "auxiliary_loss_clip": 0.01172228,
      "auxiliary_loss_mlp": 0.01030811,
      "balance_loss_clip": 1.05042708,
      "balance_loss_mlp": 1.02331889,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 2.181842919876161,
      "language_loss": 0.8137095,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83573991,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.817945957183838
    },
    {
      "auxiliary_loss_clip": 0.01163977,
      "auxiliary_loss_mlp": 0.01029913,
      "balance_loss_clip": 1.00895977,
      "balance_loss_mlp": 1.02236092,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 1.9270049485293728,
      "language_loss": 0.67014289,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69208181,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 2.646085262298584
    },
    {
      "auxiliary_loss_clip": 0.01170016,
      "auxiliary_loss_mlp": 0.01028094,
      "balance_loss_clip": 0.97438502,
      "balance_loss_mlp": 1.02072096,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 1.7227534438790282,
      "language_loss": 0.79726249,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.81924355,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.7368836402893066
    },
    {
      "auxiliary_loss_clip": 0.01176368,
      "auxiliary_loss_mlp": 0.01028265,
      "balance_loss_clip": 0.93544108,
      "balance_loss_mlp": 1.02018237,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 1.9497867597328198,
      "language_loss": 0.76729333,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.78933966,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 2.7778615951538086
    },
    {
      "auxiliary_loss_clip": 0.01179037,
      "auxiliary_loss_mlp": 0.01027985,
      "balance_loss_clip": 1.01338065,
      "balance_loss_mlp": 1.02014148,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 1.7649878468316567,
      "language_loss": 0.76374108,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.78581131,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 2.6397652626037598
    },
    {
      "auxiliary_loss_clip": 0.01166804,
      "auxiliary_loss_mlp": 0.01024919,
      "balance_loss_clip": 0.97525072,
      "balance_loss_mlp": 1.01718819,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 1.8968292601172938,
      "language_loss": 0.78793406,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.80985129,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.651608943939209
    },
    {
      "auxiliary_loss_clip": 0.01174439,
      "auxiliary_loss_mlp": 0.01030572,
      "balance_loss_clip": 0.85713589,
      "balance_loss_mlp": 1.02242422,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 1.7498905799532938,
      "language_loss": 0.77220798,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79425806,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 2.789574384689331
    },
    {
      "auxiliary_loss_clip": 0.01172417,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.01188636,
      "balance_loss_mlp": 1.02030063,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 2.137418245211094,
      "language_loss": 0.75883949,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.78084028,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.647726535797119
    },
    {
      "auxiliary_loss_clip": 0.01172729,
      "auxiliary_loss_mlp": 0.01025523,
      "balance_loss_clip": 0.97252423,
      "balance_loss_mlp": 1.01849294,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 1.8134179821541025,
      "language_loss": 0.78317171,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.8051542,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.7223315238952637
    },
    {
      "auxiliary_loss_clip": 0.01171675,
      "auxiliary_loss_mlp": 0.01025338,
      "balance_loss_clip": 1.01214814,
      "balance_loss_mlp": 1.01746464,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 2.3851652263784526,
      "language_loss": 0.75821602,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78018618,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 2.6787874698638916
    },
    {
      "auxiliary_loss_clip": 0.01170732,
      "auxiliary_loss_mlp": 0.01033077,
      "balance_loss_clip": 0.97117817,
      "balance_loss_mlp": 1.02542388,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 2.0187484817648556,
      "language_loss": 0.84335178,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.86538988,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.706834554672241
    },
    {
      "auxiliary_loss_clip": 0.01174875,
      "auxiliary_loss_mlp": 0.01029811,
      "balance_loss_clip": 1.01572776,
      "balance_loss_mlp": 1.02227736,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 1.761414331891317,
      "language_loss": 0.83012491,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.85217178,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 2.7079732418060303
    },
    {
      "auxiliary_loss_clip": 0.01170113,
      "auxiliary_loss_mlp": 0.0112337,
      "balance_loss_clip": 1.01227474,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.544453035205309,
      "language_loss": 0.69665003,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.71958488,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 2.623210906982422
    },
    {
      "auxiliary_loss_clip": 0.01170908,
      "auxiliary_loss_mlp": 0.01030455,
      "balance_loss_clip": 0.93590885,
      "balance_loss_mlp": 1.02277493,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 1.616495871596787,
      "language_loss": 0.74228513,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.7642988,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.7434604167938232
    },
    {
      "auxiliary_loss_clip": 0.01175793,
      "auxiliary_loss_mlp": 0.01024552,
      "balance_loss_clip": 1.05280364,
      "balance_loss_mlp": 1.0175488,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.14511276456334,
      "language_loss": 0.7270304,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.74903387,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.569561719894409
    },
    {
      "auxiliary_loss_clip": 0.01174682,
      "auxiliary_loss_mlp": 0.01026776,
      "balance_loss_clip": 1.01083374,
      "balance_loss_mlp": 1.01875329,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 3.0373852847119087,
      "language_loss": 0.79210675,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.81412125,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 3.586256265640259
    },
    {
      "auxiliary_loss_clip": 0.01168349,
      "auxiliary_loss_mlp": 0.0112265,
      "balance_loss_clip": 1.00957263,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 2.0782923500456394,
      "language_loss": 0.81571805,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.83862805,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.6399972438812256
    },
    {
      "auxiliary_loss_clip": 0.01181526,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.01331186,
      "balance_loss_mlp": 1.02347088,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 2.0809755521659508,
      "language_loss": 0.81979978,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.84192598,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.5844991207122803
    },
    {
      "auxiliary_loss_clip": 0.01172782,
      "auxiliary_loss_mlp": 0.01031591,
      "balance_loss_clip": 1.05127931,
      "balance_loss_mlp": 1.02430701,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 1.8298169631388037,
      "language_loss": 0.63500261,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65704632,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 3.6176435947418213
    },
    {
      "auxiliary_loss_clip": 0.01171184,
      "auxiliary_loss_mlp": 0.01027503,
      "balance_loss_clip": 1.01109672,
      "balance_loss_mlp": 1.02061892,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.5150730894332431,
      "language_loss": 0.69780231,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.71978915,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.618736743927002
    },
    {
      "auxiliary_loss_clip": 0.01178224,
      "auxiliary_loss_mlp": 0.01122881,
      "balance_loss_clip": 1.05243158,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 2.025063938039886,
      "language_loss": 0.68062866,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.70363969,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.735722064971924
    },
    {
      "auxiliary_loss_clip": 0.01182771,
      "auxiliary_loss_mlp": 0.01032394,
      "balance_loss_clip": 0.86101687,
      "balance_loss_mlp": 1.02417469,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 2.0296453682517424,
      "language_loss": 0.78102446,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80317616,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 3.7731754779815674
    },
    {
      "auxiliary_loss_clip": 0.01076468,
      "auxiliary_loss_mlp": 0.01002676,
      "balance_loss_clip": 0.94320047,
      "balance_loss_mlp": 1.00087571,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.8027029975573141,
      "language_loss": 0.56669235,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58748376,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.2594847679138184
    },
    {
      "auxiliary_loss_clip": 0.01171067,
      "auxiliary_loss_mlp": 0.01023347,
      "balance_loss_clip": 1.012923,
      "balance_loss_mlp": 1.01570547,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 1.573595049210598,
      "language_loss": 0.76066566,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78260982,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.707627058029175
    },
    {
      "auxiliary_loss_clip": 0.01169763,
      "auxiliary_loss_mlp": 0.01029004,
      "balance_loss_clip": 0.9365536,
      "balance_loss_mlp": 1.02179217,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 1.8712136846649232,
      "language_loss": 0.78704667,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.80903435,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.6990373134613037
    },
    {
      "auxiliary_loss_clip": 0.01179301,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 0.93510854,
      "balance_loss_mlp": 1.02435136,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 1.7703027923883383,
      "language_loss": 0.80015171,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.82226461,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 3.6738758087158203
    },
    {
      "auxiliary_loss_clip": 0.01174758,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.01491427,
      "balance_loss_mlp": 1.02081251,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 1.7892911705115928,
      "language_loss": 0.70939314,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.73142546,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.6949074268341064
    },
    {
      "auxiliary_loss_clip": 0.0117405,
      "auxiliary_loss_mlp": 0.01030047,
      "balance_loss_clip": 0.97666883,
      "balance_loss_mlp": 1.02256691,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.4757144852837967,
      "language_loss": 0.72184157,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.7438826,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 2.750643253326416
    },
    {
      "auxiliary_loss_clip": 0.01171517,
      "auxiliary_loss_mlp": 0.01032051,
      "balance_loss_clip": 0.97116023,
      "balance_loss_mlp": 1.02435005,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 2.1452586585551563,
      "language_loss": 0.7143966,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73643219,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.6868388652801514
    },
    {
      "auxiliary_loss_clip": 0.01173556,
      "auxiliary_loss_mlp": 0.01026618,
      "balance_loss_clip": 1.05025887,
      "balance_loss_mlp": 1.01877356,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 2.002594822957731,
      "language_loss": 0.84062684,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.86262858,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 2.6263620853424072
    },
    {
      "auxiliary_loss_clip": 0.01168615,
      "auxiliary_loss_mlp": 0.01024429,
      "balance_loss_clip": 0.96791255,
      "balance_loss_mlp": 1.01652527,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.2781372197953322,
      "language_loss": 0.68428516,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70621562,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.6318469047546387
    },
    {
      "auxiliary_loss_clip": 0.01079425,
      "auxiliary_loss_mlp": 0.01006221,
      "balance_loss_clip": 0.9425801,
      "balance_loss_mlp": 1.00451648,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7781516619115009,
      "language_loss": 0.5961042,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61696064,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 3.3622214794158936
    },
    {
      "auxiliary_loss_clip": 0.01166431,
      "auxiliary_loss_mlp": 0.01024405,
      "balance_loss_clip": 1.00859594,
      "balance_loss_mlp": 1.01673973,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 2.639770390742843,
      "language_loss": 0.86285943,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.88476777,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.678603172302246
    },
    {
      "auxiliary_loss_clip": 0.01162921,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 0.97215116,
      "balance_loss_mlp": 1.01783514,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 1.7395491217210668,
      "language_loss": 0.73000067,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.75188673,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.732184410095215
    },
    {
      "auxiliary_loss_clip": 0.01177811,
      "auxiliary_loss_mlp": 0.01028216,
      "balance_loss_clip": 1.01138854,
      "balance_loss_mlp": 1.02008295,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 2.0303853295577254,
      "language_loss": 0.6159929,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63805318,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.6508381366729736
    },
    {
      "auxiliary_loss_clip": 0.01174478,
      "auxiliary_loss_mlp": 0.01028979,
      "balance_loss_clip": 1.01401544,
      "balance_loss_mlp": 1.02140284,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 2.0192520074694107,
      "language_loss": 0.75369871,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.77573329,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 2.952866792678833
    },
    {
      "auxiliary_loss_clip": 0.01180873,
      "auxiliary_loss_mlp": 0.01029587,
      "balance_loss_clip": 0.97400838,
      "balance_loss_mlp": 1.02110517,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 1.9156488545482662,
      "language_loss": 0.71724463,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.73934919,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 2.6926910877227783
    },
    {
      "auxiliary_loss_clip": 0.0116872,
      "auxiliary_loss_mlp": 0.0102632,
      "balance_loss_clip": 0.97404706,
      "balance_loss_mlp": 1.01898813,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.0660098720944013,
      "language_loss": 0.7332561,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.75520647,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 2.628932237625122
    },
    {
      "auxiliary_loss_clip": 0.01179232,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.05371499,
      "balance_loss_mlp": 1.01966596,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 4.855982603000018,
      "language_loss": 0.8435477,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.86561191,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 2.5817058086395264
    },
    {
      "auxiliary_loss_clip": 0.01177178,
      "auxiliary_loss_mlp": 0.01034981,
      "balance_loss_clip": 1.01518893,
      "balance_loss_mlp": 1.02685714,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 1.7228905660213547,
      "language_loss": 0.81691885,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.83904046,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 2.653822898864746
    },
    {
      "auxiliary_loss_clip": 0.01176265,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.05205536,
      "balance_loss_mlp": 1.0188992,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 3.7619458125121823,
      "language_loss": 0.62113959,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.64317024,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.645946979522705
    },
    {
      "auxiliary_loss_clip": 0.01171855,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 1.01153111,
      "balance_loss_mlp": 1.01695132,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 1.9726075761564639,
      "language_loss": 0.80418289,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82614499,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 3.6266820430755615
    },
    {
      "auxiliary_loss_clip": 0.01177787,
      "auxiliary_loss_mlp": 0.01030292,
      "balance_loss_clip": 0.97788107,
      "balance_loss_mlp": 1.02284145,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 4.480903504685959,
      "language_loss": 0.68641621,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.70849699,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.6368956565856934
    },
    {
      "auxiliary_loss_clip": 0.0108237,
      "auxiliary_loss_mlp": 0.01000021,
      "balance_loss_clip": 0.90734887,
      "balance_loss_mlp": 0.99817336,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.9012466629027077,
      "language_loss": 0.61060077,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63142473,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.287855386734009
    },
    {
      "auxiliary_loss_clip": 0.01168011,
      "auxiliary_loss_mlp": 0.01030685,
      "balance_loss_clip": 0.97284257,
      "balance_loss_mlp": 1.02280569,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 2.186398921970929,
      "language_loss": 0.74483407,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76682103,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 3.5977554321289062
    },
    {
      "auxiliary_loss_clip": 0.01173376,
      "auxiliary_loss_mlp": 0.01022114,
      "balance_loss_clip": 0.97795689,
      "balance_loss_mlp": 1.01471996,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 1.8555698712257727,
      "language_loss": 0.85151613,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.87347102,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.645084857940674
    },
    {
      "auxiliary_loss_clip": 0.0117478,
      "auxiliary_loss_mlp": 0.01030216,
      "balance_loss_clip": 0.9769029,
      "balance_loss_mlp": 1.02219319,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.753641104145318,
      "language_loss": 0.76692724,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.78897715,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.7102112770080566
    },
    {
      "auxiliary_loss_clip": 0.01079088,
      "auxiliary_loss_mlp": 0.00999598,
      "balance_loss_clip": 0.94425237,
      "balance_loss_mlp": 0.99789345,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7115153384491768,
      "language_loss": 0.54760683,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.56839365,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.9959428310394287
    },
    {
      "auxiliary_loss_clip": 0.0117477,
      "auxiliary_loss_mlp": 0.0102365,
      "balance_loss_clip": 1.012429,
      "balance_loss_mlp": 1.01620507,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 1.7917849875661171,
      "language_loss": 0.79138231,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81336647,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.702059507369995
    },
    {
      "auxiliary_loss_clip": 0.01175416,
      "auxiliary_loss_mlp": 0.01024686,
      "balance_loss_clip": 1.01315856,
      "balance_loss_mlp": 1.01742053,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 1.8910983570754336,
      "language_loss": 0.77232784,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79432887,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.6586875915527344
    },
    {
      "auxiliary_loss_clip": 0.0116668,
      "auxiliary_loss_mlp": 0.01023423,
      "balance_loss_clip": 1.00874889,
      "balance_loss_mlp": 1.01601982,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.036378245139231,
      "language_loss": 0.892353,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.91425401,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.6670329570770264
    },
    {
      "auxiliary_loss_clip": 0.01088488,
      "auxiliary_loss_mlp": 0.00998943,
      "balance_loss_clip": 0.86830163,
      "balance_loss_mlp": 0.99710697,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9115170751515826,
      "language_loss": 0.63195902,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65283334,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 4.1379547119140625
    },
    {
      "auxiliary_loss_clip": 0.01165615,
      "auxiliary_loss_mlp": 0.01021902,
      "balance_loss_clip": 1.00965476,
      "balance_loss_mlp": 1.01380169,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.211914814669022,
      "language_loss": 0.79057497,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.81245011,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.6724090576171875
    },
    {
      "auxiliary_loss_clip": 0.0117035,
      "auxiliary_loss_mlp": 0.01025162,
      "balance_loss_clip": 1.01190114,
      "balance_loss_mlp": 1.01713967,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 2.658302068666858,
      "language_loss": 0.8107655,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.83272064,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.718466281890869
    },
    {
      "auxiliary_loss_clip": 0.01174648,
      "auxiliary_loss_mlp": 0.0102716,
      "balance_loss_clip": 1.01111603,
      "balance_loss_mlp": 1.01955783,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 2.119936479137371,
      "language_loss": 0.72645921,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.74847734,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 2.6943249702453613
    },
    {
      "auxiliary_loss_clip": 0.01164721,
      "auxiliary_loss_mlp": 0.01029959,
      "balance_loss_clip": 0.93258655,
      "balance_loss_mlp": 1.02261603,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 2.499414839622862,
      "language_loss": 0.70023835,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.72218513,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 2.687415361404419
    },
    {
      "auxiliary_loss_clip": 0.01177284,
      "auxiliary_loss_mlp": 0.01025195,
      "balance_loss_clip": 1.01454556,
      "balance_loss_mlp": 1.01806629,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.7016575794454158,
      "language_loss": 0.77282083,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79484558,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.712481737136841
    },
    {
      "auxiliary_loss_clip": 0.01154524,
      "auxiliary_loss_mlp": 0.01031769,
      "balance_loss_clip": 0.96831685,
      "balance_loss_mlp": 1.02340627,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 3.4429400361254228,
      "language_loss": 0.82008183,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84194475,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 2.6638081073760986
    },
    {
      "auxiliary_loss_clip": 0.01172022,
      "auxiliary_loss_mlp": 0.01024982,
      "balance_loss_clip": 1.00989532,
      "balance_loss_mlp": 1.01711392,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 2.002181134347284,
      "language_loss": 0.77845025,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.80042028,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.686345100402832
    },
    {
      "auxiliary_loss_clip": 0.01170664,
      "auxiliary_loss_mlp": 0.01020167,
      "balance_loss_clip": 0.97547072,
      "balance_loss_mlp": 1.01316667,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 2.0210765382351337,
      "language_loss": 0.71409905,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.73600733,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.655482292175293
    },
    {
      "auxiliary_loss_clip": 0.01176664,
      "auxiliary_loss_mlp": 0.01032853,
      "balance_loss_clip": 1.05090475,
      "balance_loss_mlp": 1.0250901,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 1.7690071086501822,
      "language_loss": 0.83631378,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.85840893,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 2.6222214698791504
    },
    {
      "auxiliary_loss_clip": 0.01174747,
      "auxiliary_loss_mlp": 0.01026349,
      "balance_loss_clip": 1.01099849,
      "balance_loss_mlp": 1.01848137,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 1.7805855667948187,
      "language_loss": 0.82247519,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84448612,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.6874799728393555
    },
    {
      "auxiliary_loss_clip": 0.01175785,
      "auxiliary_loss_mlp": 0.01029958,
      "balance_loss_clip": 1.05169845,
      "balance_loss_mlp": 1.02269852,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 2.111687286769386,
      "language_loss": 0.80704367,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.82910115,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 2.63234543800354
    },
    {
      "auxiliary_loss_clip": 0.0116293,
      "auxiliary_loss_mlp": 0.01023797,
      "balance_loss_clip": 0.93299156,
      "balance_loss_mlp": 1.01620901,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 1.423667122739211,
      "language_loss": 0.6676482,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68951547,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 2.7524521350860596
    },
    {
      "auxiliary_loss_clip": 0.01171584,
      "auxiliary_loss_mlp": 0.01122692,
      "balance_loss_clip": 1.01153743,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.7640078089053581,
      "language_loss": 0.75611454,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.77905738,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 2.6680972576141357
    },
    {
      "auxiliary_loss_clip": 0.01176707,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 0.97401178,
      "balance_loss_mlp": 1.01905513,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 1.6633345962862682,
      "language_loss": 0.81927705,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84131372,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 2.795771837234497
    },
    {
      "auxiliary_loss_clip": 0.01168035,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.01253569,
      "balance_loss_mlp": 1.01867628,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.4384673745034138,
      "language_loss": 0.7432307,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.76517272,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 3.703347682952881
    },
    {
      "auxiliary_loss_clip": 0.01168304,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.01244831,
      "balance_loss_mlp": 1.01846814,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 1.6425765070288223,
      "language_loss": 0.77836245,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.8003,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.677619218826294
    },
    {
      "auxiliary_loss_clip": 0.01158601,
      "auxiliary_loss_mlp": 0.01026593,
      "balance_loss_clip": 0.89487398,
      "balance_loss_mlp": 1.0193454,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 1.6462647274254882,
      "language_loss": 0.75904095,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78089291,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.878448963165283
    },
    {
      "auxiliary_loss_clip": 0.01177938,
      "auxiliary_loss_mlp": 0.01027239,
      "balance_loss_clip": 0.97327268,
      "balance_loss_mlp": 1.01949668,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 2.3213480082713365,
      "language_loss": 0.72931707,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.75136882,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.681361675262451
    },
    {
      "auxiliary_loss_clip": 0.01171943,
      "auxiliary_loss_mlp": 0.01029649,
      "balance_loss_clip": 0.9774847,
      "balance_loss_mlp": 1.02234721,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 5.586758450630948,
      "language_loss": 0.91707987,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.93909574,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 3.600480794906616
    },
    {
      "auxiliary_loss_clip": 0.01163226,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 0.97327632,
      "balance_loss_mlp": 1.01918042,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 3.7740869082633,
      "language_loss": 0.66552258,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.68741846,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.628237724304199
    },
    {
      "auxiliary_loss_clip": 0.01168864,
      "auxiliary_loss_mlp": 0.01031756,
      "balance_loss_clip": 0.89476097,
      "balance_loss_mlp": 1.02372718,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 1.9167733343700302,
      "language_loss": 0.76260805,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78461432,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.791165351867676
    },
    {
      "auxiliary_loss_clip": 0.01167855,
      "auxiliary_loss_mlp": 0.01023553,
      "balance_loss_clip": 0.97638559,
      "balance_loss_mlp": 1.0158937,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 2.1834545761605897,
      "language_loss": 0.79132092,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.81323504,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 3.6856863498687744
    },
    {
      "auxiliary_loss_clip": 0.01176604,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.05290532,
      "balance_loss_mlp": 1.0181371,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 1.6852066697785448,
      "language_loss": 0.72874498,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.75076997,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.6446151733398438
    },
    {
      "auxiliary_loss_clip": 0.01173385,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.01322424,
      "balance_loss_mlp": 1.0176158,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 1.6886718177087907,
      "language_loss": 0.72032708,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74231374,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.760829210281372
    },
    {
      "auxiliary_loss_clip": 0.01174795,
      "auxiliary_loss_mlp": 0.01031412,
      "balance_loss_clip": 0.93624824,
      "balance_loss_mlp": 1.0242182,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 1.583589444622084,
      "language_loss": 0.84954679,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.87160885,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.765378475189209
    },
    {
      "auxiliary_loss_clip": 0.01176797,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.05089355,
      "balance_loss_mlp": 1.02027678,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 1.487054163134341,
      "language_loss": 0.75243551,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77448261,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 3.5516374111175537
    },
    {
      "auxiliary_loss_clip": 0.0116591,
      "auxiliary_loss_mlp": 0.01022691,
      "balance_loss_clip": 0.97146618,
      "balance_loss_mlp": 1.01529348,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.71248508188343,
      "language_loss": 0.64732778,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66921377,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.858273983001709
    },
    {
      "auxiliary_loss_clip": 0.01167742,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 0.89434886,
      "balance_loss_mlp": 1.0229013,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 2.1477416311648985,
      "language_loss": 0.7614249,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.78340816,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.7632627487182617
    },
    {
      "auxiliary_loss_clip": 0.01175972,
      "auxiliary_loss_mlp": 0.01032927,
      "balance_loss_clip": 0.97553003,
      "balance_loss_mlp": 1.0255239,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.0414016687088945,
      "language_loss": 0.80030489,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.82239389,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 2.7199714183807373
    },
    {
      "auxiliary_loss_clip": 0.01174496,
      "auxiliary_loss_mlp": 0.01031695,
      "balance_loss_clip": 1.01288664,
      "balance_loss_mlp": 1.02318335,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 1.9057437769198378,
      "language_loss": 0.73677266,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.7588346,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.629354238510132
    },
    {
      "auxiliary_loss_clip": 0.01173172,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 0.97619104,
      "balance_loss_mlp": 1.01939666,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 2.2275917344385534,
      "language_loss": 0.7700516,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.7920537,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 2.714325189590454
    },
    {
      "auxiliary_loss_clip": 0.01069232,
      "auxiliary_loss_mlp": 0.01002572,
      "balance_loss_clip": 0.94230366,
      "balance_loss_mlp": 1.00097454,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 0.7404001513035104,
      "language_loss": 0.59854656,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.6192646,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 3.2927589416503906
    },
    {
      "auxiliary_loss_clip": 0.01176923,
      "auxiliary_loss_mlp": 0.0102926,
      "balance_loss_clip": 0.97312129,
      "balance_loss_mlp": 1.02182698,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 1.6296513958422951,
      "language_loss": 0.77403402,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79609585,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.731757402420044
    },
    {
      "auxiliary_loss_clip": 0.01075013,
      "auxiliary_loss_mlp": 0.01003953,
      "balance_loss_clip": 0.97806442,
      "balance_loss_mlp": 1.00226045,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.999325375590386,
      "language_loss": 0.68743938,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70822901,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.306774616241455
    },
    {
      "auxiliary_loss_clip": 0.01175545,
      "auxiliary_loss_mlp": 0.01028803,
      "balance_loss_clip": 0.86007249,
      "balance_loss_mlp": 1.0214057,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 1.6840304978772629,
      "language_loss": 0.82626092,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.84830439,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 2.844304323196411
    },
    {
      "auxiliary_loss_clip": 0.01172428,
      "auxiliary_loss_mlp": 0.01028648,
      "balance_loss_clip": 0.97040623,
      "balance_loss_mlp": 1.02076268,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 2.572769295949536,
      "language_loss": 0.80670482,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.82871556,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.680852174758911
    },
    {
      "auxiliary_loss_clip": 0.0117783,
      "auxiliary_loss_mlp": 0.0102572,
      "balance_loss_clip": 0.97431529,
      "balance_loss_mlp": 1.01790547,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 2.148523099456146,
      "language_loss": 0.79310155,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.81513709,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 2.699856758117676
    },
    {
      "auxiliary_loss_clip": 0.01173756,
      "auxiliary_loss_mlp": 0.01033211,
      "balance_loss_clip": 0.97308159,
      "balance_loss_mlp": 1.02552807,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 1.863849048831106,
      "language_loss": 0.81985176,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84192145,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 2.7257943153381348
    },
    {
      "auxiliary_loss_clip": 0.01174374,
      "auxiliary_loss_mlp": 0.01024822,
      "balance_loss_clip": 0.89885068,
      "balance_loss_mlp": 1.01709688,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 1.5810766195283208,
      "language_loss": 0.85530871,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.87730062,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 2.7810370922088623
    },
    {
      "auxiliary_loss_clip": 0.0107015,
      "auxiliary_loss_mlp": 0.01005171,
      "balance_loss_clip": 1.01417351,
      "balance_loss_mlp": 1.00345409,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7421974344085144,
      "language_loss": 0.55870813,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.57946134,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 3.1846373081207275
    },
    {
      "auxiliary_loss_clip": 0.01169985,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 0.93450242,
      "balance_loss_mlp": 1.01921213,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.1621083930153318,
      "language_loss": 0.66010261,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.68206608,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 3.6268954277038574
    },
    {
      "auxiliary_loss_clip": 0.01179437,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.05267906,
      "balance_loss_mlp": 1.020679,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 1.9942710892160898,
      "language_loss": 0.7483145,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.77039158,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.7123188972473145
    },
    {
      "auxiliary_loss_clip": 0.01165162,
      "auxiliary_loss_mlp": 0.0103092,
      "balance_loss_clip": 0.90124357,
      "balance_loss_mlp": 1.02308226,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 1.9988902246332048,
      "language_loss": 0.81699502,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.83895588,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.8030877113342285
    },
    {
      "auxiliary_loss_clip": 0.0116589,
      "auxiliary_loss_mlp": 0.01023064,
      "balance_loss_clip": 0.89258683,
      "balance_loss_mlp": 1.01594138,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.1176772405040163,
      "language_loss": 0.78035873,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.80224824,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.7520222663879395
    },
    {
      "auxiliary_loss_clip": 0.01172774,
      "auxiliary_loss_mlp": 0.01023474,
      "balance_loss_clip": 1.011374,
      "balance_loss_mlp": 1.01562691,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 4.593705949270212,
      "language_loss": 0.68472016,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.70668262,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 3.657846689224243
    },
    {
      "auxiliary_loss_clip": 0.01163097,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 0.9704054,
      "balance_loss_mlp": 1.02096415,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 2.148057278187096,
      "language_loss": 0.83021867,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85213202,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.7036218643188477
    },
    {
      "auxiliary_loss_clip": 0.01160445,
      "auxiliary_loss_mlp": 0.01022747,
      "balance_loss_clip": 0.93493599,
      "balance_loss_mlp": 1.01524234,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 1.8122234543919784,
      "language_loss": 0.78400767,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.80583954,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.7622835636138916
    },
    {
      "auxiliary_loss_clip": 0.01165415,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 0.93199652,
      "balance_loss_mlp": 1.01934361,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 1.7477053568224366,
      "language_loss": 0.78596908,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.80789387,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 3.5730464458465576
    },
    {
      "auxiliary_loss_clip": 0.01174396,
      "auxiliary_loss_mlp": 0.01026468,
      "balance_loss_clip": 1.04899859,
      "balance_loss_mlp": 1.01861262,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.1093368140561473,
      "language_loss": 0.79129207,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.81330067,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.624187469482422
    },
    {
      "auxiliary_loss_clip": 0.01176591,
      "auxiliary_loss_mlp": 0.01029795,
      "balance_loss_clip": 1.01372623,
      "balance_loss_mlp": 1.02204633,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.7410923100790834,
      "language_loss": 0.67068863,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.69275248,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.6919617652893066
    },
    {
      "auxiliary_loss_clip": 0.01150053,
      "auxiliary_loss_mlp": 0.01024591,
      "balance_loss_clip": 0.85747713,
      "balance_loss_mlp": 1.01691973,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 1.6234860811729417,
      "language_loss": 0.73883748,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.760584,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 3.7288870811462402
    },
    {
      "auxiliary_loss_clip": 0.01160485,
      "auxiliary_loss_mlp": 0.01027762,
      "balance_loss_clip": 0.9315573,
      "balance_loss_mlp": 1.01951861,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 2.8633053061204286,
      "language_loss": 0.76189291,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.78377545,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.732875347137451
    },
    {
      "auxiliary_loss_clip": 0.0115693,
      "auxiliary_loss_mlp": 0.01024137,
      "balance_loss_clip": 0.97036564,
      "balance_loss_mlp": 1.01688945,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.3894788516760102,
      "language_loss": 0.70700735,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.72881806,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.738770008087158
    },
    {
      "auxiliary_loss_clip": 0.01172918,
      "auxiliary_loss_mlp": 0.01023671,
      "balance_loss_clip": 0.93473047,
      "balance_loss_mlp": 1.01582718,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 4.407061538691369,
      "language_loss": 0.69437271,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.71633863,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.6807284355163574
    },
    {
      "auxiliary_loss_clip": 0.01174314,
      "auxiliary_loss_mlp": 0.01027306,
      "balance_loss_clip": 1.05232072,
      "balance_loss_mlp": 1.02009392,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.6104928706774648,
      "language_loss": 0.72290313,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.7449193,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.669229507446289
    },
    {
      "auxiliary_loss_clip": 0.01178573,
      "auxiliary_loss_mlp": 0.01027987,
      "balance_loss_clip": 0.97255075,
      "balance_loss_mlp": 1.02017307,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 2.6952107407727146,
      "language_loss": 0.83023071,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.85229635,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 2.791064739227295
    },
    {
      "auxiliary_loss_clip": 0.01169692,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.01390123,
      "balance_loss_mlp": 1.01993299,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 1.8787833095572182,
      "language_loss": 0.82850647,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.8504774,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.678405284881592
    },
    {
      "auxiliary_loss_clip": 0.0117375,
      "auxiliary_loss_mlp": 0.01026541,
      "balance_loss_clip": 1.01316428,
      "balance_loss_mlp": 1.01927257,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.314083876610931,
      "language_loss": 0.71678555,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.73878843,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 2.70883846282959
    },
    {
      "auxiliary_loss_clip": 0.01170677,
      "auxiliary_loss_mlp": 0.010305,
      "balance_loss_clip": 1.01406765,
      "balance_loss_mlp": 1.02323413,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 1.7086835090041654,
      "language_loss": 0.72620237,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74821413,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.761516809463501
    },
    {
      "auxiliary_loss_clip": 0.01172045,
      "auxiliary_loss_mlp": 0.0102874,
      "balance_loss_clip": 0.93634468,
      "balance_loss_mlp": 1.02166486,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 1.419723249428737,
      "language_loss": 0.74011272,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.7621206,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.7844183444976807
    },
    {
      "auxiliary_loss_clip": 0.01173195,
      "auxiliary_loss_mlp": 0.01031507,
      "balance_loss_clip": 0.93975806,
      "balance_loss_mlp": 1.02324581,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 2.3441357432506136,
      "language_loss": 0.73310649,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.75515354,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 2.677733898162842
    },
    {
      "auxiliary_loss_clip": 0.01159767,
      "auxiliary_loss_mlp": 0.01122531,
      "balance_loss_clip": 1.01045728,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 2.100187323344353,
      "language_loss": 0.83790815,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.86073112,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 2.6587533950805664
    },
    {
      "auxiliary_loss_clip": 0.01173708,
      "auxiliary_loss_mlp": 0.01028957,
      "balance_loss_clip": 1.05205595,
      "balance_loss_mlp": 1.02109528,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 1.5273493201318196,
      "language_loss": 0.70591605,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.72794271,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.6389923095703125
    },
    {
      "auxiliary_loss_clip": 0.0107523,
      "auxiliary_loss_mlp": 0.01000032,
      "balance_loss_clip": 0.90207839,
      "balance_loss_mlp": 0.99826759,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.8001991900012321,
      "language_loss": 0.61969638,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.64044905,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.348389148712158
    },
    {
      "auxiliary_loss_clip": 0.01176283,
      "auxiliary_loss_mlp": 0.01022321,
      "balance_loss_clip": 0.97368252,
      "balance_loss_mlp": 1.01479244,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.5019344876155356,
      "language_loss": 0.70847118,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.73045719,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 2.6813278198242188
    },
    {
      "auxiliary_loss_clip": 0.01176429,
      "auxiliary_loss_mlp": 0.01028252,
      "balance_loss_clip": 0.97412622,
      "balance_loss_mlp": 1.02053905,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 1.6043075690117343,
      "language_loss": 0.79136777,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81341457,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 2.696263074874878
    },
    {
      "auxiliary_loss_clip": 0.01168875,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 0.97496474,
      "balance_loss_mlp": 1.01955462,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 2.222708711526621,
      "language_loss": 0.7390312,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.76098764,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 2.770901679992676
    },
    {
      "auxiliary_loss_clip": 0.01164294,
      "auxiliary_loss_mlp": 0.01031228,
      "balance_loss_clip": 0.89837623,
      "balance_loss_mlp": 1.02334774,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 1.688696082041837,
      "language_loss": 0.79649174,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.81844693,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 3.7098007202148438
    },
    {
      "auxiliary_loss_clip": 0.01171839,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 0.97494471,
      "balance_loss_mlp": 1.01716089,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.6410677105662794,
      "language_loss": 0.70870233,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.7306729,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.6911849975585938
    },
    {
      "auxiliary_loss_clip": 0.01074474,
      "auxiliary_loss_mlp": 0.01001421,
      "balance_loss_clip": 0.94041681,
      "balance_loss_mlp": 0.99975187,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.8243586683075905,
      "language_loss": 0.60389978,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62465876,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.275149345397949
    },
    {
      "auxiliary_loss_clip": 0.01166903,
      "auxiliary_loss_mlp": 0.01025271,
      "balance_loss_clip": 1.01257575,
      "balance_loss_mlp": 1.01793408,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.5613844825268155,
      "language_loss": 0.75275981,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.77468157,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.6823980808258057
    },
    {
      "auxiliary_loss_clip": 0.01180497,
      "auxiliary_loss_mlp": 0.01036344,
      "balance_loss_clip": 0.97824252,
      "balance_loss_mlp": 1.02821934,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.8225584981048755,
      "language_loss": 0.76408833,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78625673,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 3.6022839546203613
    },
    {
      "auxiliary_loss_clip": 0.01173807,
      "auxiliary_loss_mlp": 0.01028197,
      "balance_loss_clip": 1.01384282,
      "balance_loss_mlp": 1.02078235,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.430291003350859,
      "language_loss": 0.85768527,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.87970531,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.609912395477295
    },
    {
      "auxiliary_loss_clip": 0.01163528,
      "auxiliary_loss_mlp": 0.01025193,
      "balance_loss_clip": 0.89313173,
      "balance_loss_mlp": 1.01799595,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 1.7811179777120951,
      "language_loss": 0.78110683,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80299401,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 3.7311887741088867
    },
    {
      "auxiliary_loss_clip": 0.01172526,
      "auxiliary_loss_mlp": 0.01025225,
      "balance_loss_clip": 0.93687803,
      "balance_loss_mlp": 1.01754808,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 2.741784543450186,
      "language_loss": 0.76218492,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.7841624,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.63808274269104
    },
    {
      "auxiliary_loss_clip": 0.01174214,
      "auxiliary_loss_mlp": 0.01028858,
      "balance_loss_clip": 0.8967427,
      "balance_loss_mlp": 1.02129364,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 1.6102037032520118,
      "language_loss": 0.67592037,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.69795114,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.8633482456207275
    },
    {
      "auxiliary_loss_clip": 0.011739,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 0.97751272,
      "balance_loss_mlp": 1.01657319,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 1.956902567418706,
      "language_loss": 0.66025215,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.682239,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 3.717622995376587
    },
    {
      "auxiliary_loss_clip": 0.01172674,
      "auxiliary_loss_mlp": 0.01024733,
      "balance_loss_clip": 1.01195133,
      "balance_loss_mlp": 1.01702023,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 2.1447276975347087,
      "language_loss": 0.67949605,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.70147014,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.7026333808898926
    },
    {
      "auxiliary_loss_clip": 0.01174321,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.01167679,
      "balance_loss_mlp": 1.01875544,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 1.678283771622943,
      "language_loss": 0.65292621,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.67493165,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.812150716781616
    },
    {
      "auxiliary_loss_clip": 0.01166092,
      "auxiliary_loss_mlp": 0.01029675,
      "balance_loss_clip": 1.01084888,
      "balance_loss_mlp": 1.02170563,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 2.442846494854772,
      "language_loss": 0.79537207,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.81732976,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.6285347938537598
    },
    {
      "auxiliary_loss_clip": 0.01172823,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.05152512,
      "balance_loss_mlp": 1.01809299,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 1.8488810105844748,
      "language_loss": 0.84791791,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.86989909,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.6362738609313965
    },
    {
      "auxiliary_loss_clip": 0.01174867,
      "auxiliary_loss_mlp": 0.01025309,
      "balance_loss_clip": 0.90139019,
      "balance_loss_mlp": 1.01759589,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 1.9989615431461742,
      "language_loss": 0.71057761,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.73257935,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 2.7919795513153076
    },
    {
      "auxiliary_loss_clip": 0.01171596,
      "auxiliary_loss_mlp": 0.01026669,
      "balance_loss_clip": 0.97690493,
      "balance_loss_mlp": 1.01860452,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 1.8746057721356904,
      "language_loss": 0.56850934,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59049195,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.8101730346679688
    },
    {
      "auxiliary_loss_clip": 0.0116077,
      "auxiliary_loss_mlp": 0.01028682,
      "balance_loss_clip": 0.97094542,
      "balance_loss_mlp": 1.02124953,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 2.0351558808883325,
      "language_loss": 0.86791599,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.88981056,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 2.6706180572509766
    },
    {
      "auxiliary_loss_clip": 0.01172797,
      "auxiliary_loss_mlp": 0.01038671,
      "balance_loss_clip": 0.97424316,
      "balance_loss_mlp": 1.03039765,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 12.605694393460332,
      "language_loss": 0.81779039,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.83990502,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.6634650230407715
    },
    {
      "auxiliary_loss_clip": 0.01174336,
      "auxiliary_loss_mlp": 0.0103118,
      "balance_loss_clip": 1.01557326,
      "balance_loss_mlp": 1.0237056,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 1.8488287388489455,
      "language_loss": 0.78566039,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.80771554,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 2.631263017654419
    },
    {
      "auxiliary_loss_clip": 0.01172179,
      "auxiliary_loss_mlp": 0.01025664,
      "balance_loss_clip": 1.0126791,
      "balance_loss_mlp": 1.0180881,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 1.864346494064599,
      "language_loss": 0.85336709,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87534559,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.6816160678863525
    },
    {
      "auxiliary_loss_clip": 0.01174123,
      "auxiliary_loss_mlp": 0.01027791,
      "balance_loss_clip": 1.01251638,
      "balance_loss_mlp": 1.01956582,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 2.040737834314138,
      "language_loss": 0.69746053,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.71947968,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 2.7073161602020264
    },
    {
      "auxiliary_loss_clip": 0.01173431,
      "auxiliary_loss_mlp": 0.0102581,
      "balance_loss_clip": 0.9370122,
      "balance_loss_mlp": 1.01756692,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.9693823241116069,
      "language_loss": 0.73497069,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.75696307,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.8025882244110107
    },
    {
      "auxiliary_loss_clip": 0.01170727,
      "auxiliary_loss_mlp": 0.01025939,
      "balance_loss_clip": 0.97561687,
      "balance_loss_mlp": 1.01821399,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 1.9393744449740848,
      "language_loss": 0.69038719,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71235389,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.824862480163574
    },
    {
      "auxiliary_loss_clip": 0.01173648,
      "auxiliary_loss_mlp": 0.01025462,
      "balance_loss_clip": 1.01385128,
      "balance_loss_mlp": 1.01847315,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 1.7060467627939873,
      "language_loss": 0.71983063,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74182171,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 2.6687376499176025
    },
    {
      "auxiliary_loss_clip": 0.01177118,
      "auxiliary_loss_mlp": 0.01022211,
      "balance_loss_clip": 0.93732089,
      "balance_loss_mlp": 1.01442671,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 1.9653903056684936,
      "language_loss": 0.61979723,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.64179051,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 2.761899471282959
    },
    {
      "auxiliary_loss_clip": 0.01166016,
      "auxiliary_loss_mlp": 0.0103011,
      "balance_loss_clip": 0.97777212,
      "balance_loss_mlp": 1.02301133,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 1.5365047236456009,
      "language_loss": 0.77026248,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79222375,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 3.6487390995025635
    },
    {
      "auxiliary_loss_clip": 0.01180631,
      "auxiliary_loss_mlp": 0.01027509,
      "balance_loss_clip": 0.97791505,
      "balance_loss_mlp": 1.01926541,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 1.9235046245666862,
      "language_loss": 0.73354137,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.75562274,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.6626038551330566
    },
    {
      "auxiliary_loss_clip": 0.01173976,
      "auxiliary_loss_mlp": 0.0103084,
      "balance_loss_clip": 0.93681097,
      "balance_loss_mlp": 1.02362823,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 1.7166789145351782,
      "language_loss": 0.77938247,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80143058,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.785268545150757
    },
    {
      "auxiliary_loss_clip": 0.01164716,
      "auxiliary_loss_mlp": 0.01123088,
      "balance_loss_clip": 0.97218621,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.010838180539087,
      "language_loss": 0.78247792,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.80535591,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.634241819381714
    },
    {
      "auxiliary_loss_clip": 0.01164376,
      "auxiliary_loss_mlp": 0.01033437,
      "balance_loss_clip": 0.97201455,
      "balance_loss_mlp": 1.02545619,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 2.02756106894906,
      "language_loss": 0.71555859,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.73753679,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.7450969219207764
    },
    {
      "auxiliary_loss_clip": 0.01164541,
      "auxiliary_loss_mlp": 0.0103338,
      "balance_loss_clip": 0.93396753,
      "balance_loss_mlp": 1.02581656,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 1.8289295357045885,
      "language_loss": 0.86030221,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88228142,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 3.763535737991333
    },
    {
      "auxiliary_loss_clip": 0.01169176,
      "auxiliary_loss_mlp": 0.0102594,
      "balance_loss_clip": 1.01101494,
      "balance_loss_mlp": 1.01847148,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 1.9326819437535177,
      "language_loss": 0.90535617,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.92730737,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.6137964725494385
    },
    {
      "auxiliary_loss_clip": 0.01178041,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 1.0131495,
      "balance_loss_mlp": 1.02075887,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 1.6297212502493128,
      "language_loss": 0.81600231,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83806914,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 3.5824038982391357
    },
    {
      "auxiliary_loss_clip": 0.01175946,
      "auxiliary_loss_mlp": 0.01025217,
      "balance_loss_clip": 1.05208802,
      "balance_loss_mlp": 1.01720309,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.6816300407431368,
      "language_loss": 0.87802023,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.90003181,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.638383388519287
    },
    {
      "auxiliary_loss_clip": 0.0117695,
      "auxiliary_loss_mlp": 0.01024848,
      "balance_loss_clip": 0.97474432,
      "balance_loss_mlp": 1.01699221,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.5993120857504985,
      "language_loss": 0.85962588,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.88164383,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.7295517921447754
    },
    {
      "auxiliary_loss_clip": 0.01174654,
      "auxiliary_loss_mlp": 0.01123349,
      "balance_loss_clip": 1.05057049,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 1.7633871687166438,
      "language_loss": 0.8110733,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.83405334,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 3.547368288040161
    },
    {
      "auxiliary_loss_clip": 0.01169262,
      "auxiliary_loss_mlp": 0.01026947,
      "balance_loss_clip": 0.97139376,
      "balance_loss_mlp": 1.02021503,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.5098922076866907,
      "language_loss": 0.75485927,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77682137,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.770841360092163
    },
    {
      "auxiliary_loss_clip": 0.01179361,
      "auxiliary_loss_mlp": 0.01031477,
      "balance_loss_clip": 1.01503778,
      "balance_loss_mlp": 1.02410984,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 2.0372565874455986,
      "language_loss": 0.83064777,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85275614,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.621615409851074
    },
    {
      "auxiliary_loss_clip": 0.01172297,
      "auxiliary_loss_mlp": 0.01027575,
      "balance_loss_clip": 1.01318598,
      "balance_loss_mlp": 1.02009499,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 1.9945191568043545,
      "language_loss": 0.73723733,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.75923604,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.626302719116211
    },
    {
      "auxiliary_loss_clip": 0.01163885,
      "auxiliary_loss_mlp": 0.01029905,
      "balance_loss_clip": 0.8940891,
      "balance_loss_mlp": 1.02231193,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.1103448309552437,
      "language_loss": 0.78351587,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.80545378,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.8225388526916504
    },
    {
      "auxiliary_loss_clip": 0.01174694,
      "auxiliary_loss_mlp": 0.01034902,
      "balance_loss_clip": 0.97673941,
      "balance_loss_mlp": 1.02686715,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 1.7174558803868474,
      "language_loss": 0.63150299,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.6535989,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.7110037803649902
    },
    {
      "auxiliary_loss_clip": 0.01158976,
      "auxiliary_loss_mlp": 0.01028634,
      "balance_loss_clip": 0.93254602,
      "balance_loss_mlp": 1.0210402,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 1.5535963437084315,
      "language_loss": 0.86360288,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88547897,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 2.7513442039489746
    },
    {
      "auxiliary_loss_clip": 0.01177626,
      "auxiliary_loss_mlp": 0.01026665,
      "balance_loss_clip": 0.97576386,
      "balance_loss_mlp": 1.01908028,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.6109983527383334,
      "language_loss": 0.74700129,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.76904416,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 2.6921310424804688
    },
    {
      "auxiliary_loss_clip": 0.01172498,
      "auxiliary_loss_mlp": 0.01025093,
      "balance_loss_clip": 1.00992095,
      "balance_loss_mlp": 1.01677775,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 1.8373559273573667,
      "language_loss": 0.77194774,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.79392362,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 2.744241237640381
    },
    {
      "auxiliary_loss_clip": 0.01174679,
      "auxiliary_loss_mlp": 0.01024821,
      "balance_loss_clip": 0.97492939,
      "balance_loss_mlp": 1.01703691,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 1.7925732366542797,
      "language_loss": 0.73588884,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.75788379,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 2.8181698322296143
    },
    {
      "auxiliary_loss_clip": 0.01173451,
      "auxiliary_loss_mlp": 0.01028156,
      "balance_loss_clip": 1.01336002,
      "balance_loss_mlp": 1.02031767,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 1.7179898968276788,
      "language_loss": 0.89358699,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91560304,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 2.6486005783081055
    },
    {
      "auxiliary_loss_clip": 0.01172076,
      "auxiliary_loss_mlp": 0.0102723,
      "balance_loss_clip": 1.01252174,
      "balance_loss_mlp": 1.01994371,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 1.8064176760110027,
      "language_loss": 0.75111997,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77311301,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.683978319168091
    },
    {
      "auxiliary_loss_clip": 0.01076321,
      "auxiliary_loss_mlp": 0.01116749,
      "balance_loss_clip": 0.90323544,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7576528681290013,
      "language_loss": 0.58066475,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.60259545,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.2089178562164307
    },
    {
      "auxiliary_loss_clip": 0.0117232,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.01184225,
      "balance_loss_mlp": 1.01834035,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 1.9065628710219116,
      "language_loss": 0.68909943,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.71108627,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.6710429191589355
    },
    {
      "auxiliary_loss_clip": 0.01164377,
      "auxiliary_loss_mlp": 0.01030553,
      "balance_loss_clip": 1.01108313,
      "balance_loss_mlp": 1.02363276,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.6906500363476953,
      "language_loss": 0.84352171,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86547107,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 2.6360442638397217
    },
    {
      "auxiliary_loss_clip": 0.01174188,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 0.93869764,
      "balance_loss_mlp": 1.02100313,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 1.976383219402723,
      "language_loss": 0.73688126,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.75891489,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 2.6946823596954346
    },
    {
      "auxiliary_loss_clip": 0.01170596,
      "auxiliary_loss_mlp": 0.0102406,
      "balance_loss_clip": 1.04954839,
      "balance_loss_mlp": 1.01700246,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.010467377041633,
      "language_loss": 0.71476889,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73671538,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 3.569782018661499
    },
    {
      "auxiliary_loss_clip": 0.01176534,
      "auxiliary_loss_mlp": 0.01028439,
      "balance_loss_clip": 1.01108801,
      "balance_loss_mlp": 1.02069354,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.4992809212535814,
      "language_loss": 0.62003148,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64208126,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.83721661567688
    },
    {
      "auxiliary_loss_clip": 0.0116303,
      "auxiliary_loss_mlp": 0.01031362,
      "balance_loss_clip": 0.93222332,
      "balance_loss_mlp": 1.02372646,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.8115348989296236,
      "language_loss": 0.68378878,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.7057327,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.7070794105529785
    },
    {
      "auxiliary_loss_clip": 0.01177784,
      "auxiliary_loss_mlp": 0.01027271,
      "balance_loss_clip": 1.05318367,
      "balance_loss_mlp": 1.01960588,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 1.7395487130095713,
      "language_loss": 0.70865446,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73070502,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.687243938446045
    },
    {
      "auxiliary_loss_clip": 0.01165822,
      "auxiliary_loss_mlp": 0.01027321,
      "balance_loss_clip": 1.01031983,
      "balance_loss_mlp": 1.01974857,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.186316907706923,
      "language_loss": 0.80803216,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.82996356,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.665675163269043
    },
    {
      "auxiliary_loss_clip": 0.0116341,
      "auxiliary_loss_mlp": 0.01023038,
      "balance_loss_clip": 0.96970308,
      "balance_loss_mlp": 1.01540279,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 1.5481445788396035,
      "language_loss": 0.71270812,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73457265,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 3.5567944049835205
    },
    {
      "auxiliary_loss_clip": 0.01172673,
      "auxiliary_loss_mlp": 0.01023446,
      "balance_loss_clip": 1.01193631,
      "balance_loss_mlp": 1.01639771,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.6578809473770688,
      "language_loss": 0.79563934,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.81760055,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.8790242671966553
    },
    {
      "auxiliary_loss_clip": 0.01164709,
      "auxiliary_loss_mlp": 0.01031099,
      "balance_loss_clip": 0.9703114,
      "balance_loss_mlp": 1.02322555,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 2.1163780067994002,
      "language_loss": 0.90676647,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.92872459,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.677145004272461
    },
    {
      "auxiliary_loss_clip": 0.01173907,
      "auxiliary_loss_mlp": 0.01032476,
      "balance_loss_clip": 0.97383773,
      "balance_loss_mlp": 1.02516222,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 1.9935844601329589,
      "language_loss": 0.79668415,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.81874794,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 3.611171007156372
    },
    {
      "auxiliary_loss_clip": 0.01176322,
      "auxiliary_loss_mlp": 0.01027432,
      "balance_loss_clip": 0.97370976,
      "balance_loss_mlp": 1.02073205,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 1.816667868388869,
      "language_loss": 0.83301526,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.85505283,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 3.5683324337005615
    },
    {
      "auxiliary_loss_clip": 0.01177191,
      "auxiliary_loss_mlp": 0.01031077,
      "balance_loss_clip": 1.05296552,
      "balance_loss_mlp": 1.0235815,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.533825461020709,
      "language_loss": 0.83138001,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.8534627,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.570295572280884
    },
    {
      "auxiliary_loss_clip": 0.01176983,
      "auxiliary_loss_mlp": 0.01034964,
      "balance_loss_clip": 1.01167476,
      "balance_loss_mlp": 1.02736461,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.7992782217281755,
      "language_loss": 0.79472244,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.81684196,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.660071611404419
    },
    {
      "auxiliary_loss_clip": 0.01169056,
      "auxiliary_loss_mlp": 0.01022816,
      "balance_loss_clip": 1.01085091,
      "balance_loss_mlp": 1.01571429,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 1.931845398880066,
      "language_loss": 0.72281003,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.7447288,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.5925042629241943
    },
    {
      "auxiliary_loss_clip": 0.01174947,
      "auxiliary_loss_mlp": 0.01123788,
      "balance_loss_clip": 0.93617356,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 1.5071514471523701,
      "language_loss": 0.74481797,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.76780534,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.90096378326416
    },
    {
      "auxiliary_loss_clip": 0.01172415,
      "auxiliary_loss_mlp": 0.01027671,
      "balance_loss_clip": 1.01283288,
      "balance_loss_mlp": 1.0201602,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 1.7883438721789156,
      "language_loss": 0.73852217,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.76052302,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 2.6896591186523438
    },
    {
      "auxiliary_loss_clip": 0.01169906,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 0.97692597,
      "balance_loss_mlp": 1.01794839,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 2.3550460786378427,
      "language_loss": 0.78467172,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.8066209,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.676488161087036
    },
    {
      "auxiliary_loss_clip": 0.01156937,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 0.97067153,
      "balance_loss_mlp": 1.0195744,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 2.1907880252530094,
      "language_loss": 0.71455336,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73639292,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.721961498260498
    },
    {
      "auxiliary_loss_clip": 0.01177729,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 0.97576761,
      "balance_loss_mlp": 1.01780057,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 1.8835736790974649,
      "language_loss": 0.76604694,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.78807533,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 2.6390323638916016
    },
    {
      "auxiliary_loss_clip": 0.0117507,
      "auxiliary_loss_mlp": 0.01024168,
      "balance_loss_clip": 1.05262744,
      "balance_loss_mlp": 1.0163126,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 1.7579351546201778,
      "language_loss": 0.70133483,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.72332716,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 2.686959743499756
    },
    {
      "auxiliary_loss_clip": 0.01175183,
      "auxiliary_loss_mlp": 0.01023898,
      "balance_loss_clip": 0.9729054,
      "balance_loss_mlp": 1.01696265,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.6876599165559552,
      "language_loss": 0.78231895,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80430979,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 2.7022931575775146
    },
    {
      "auxiliary_loss_clip": 0.01172827,
      "auxiliary_loss_mlp": 0.01027985,
      "balance_loss_clip": 0.89813596,
      "balance_loss_mlp": 1.02026057,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.1326507520562847,
      "language_loss": 0.76118702,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78319514,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 2.680922508239746
    },
    {
      "auxiliary_loss_clip": 0.01174717,
      "auxiliary_loss_mlp": 0.01025343,
      "balance_loss_clip": 1.05049455,
      "balance_loss_mlp": 1.01776135,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 1.984952051394037,
      "language_loss": 0.75937259,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.7813732,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.674656391143799
    },
    {
      "auxiliary_loss_clip": 0.01165671,
      "auxiliary_loss_mlp": 0.0102876,
      "balance_loss_clip": 0.97284174,
      "balance_loss_mlp": 1.02123153,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.8394222113279566,
      "language_loss": 0.69731832,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.71926266,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 2.7485368251800537
    },
    {
      "auxiliary_loss_clip": 0.01171258,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.01611555,
      "balance_loss_mlp": 1.01824021,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 1.6390237952245776,
      "language_loss": 0.76762414,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78958833,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.6090774536132812
    },
    {
      "auxiliary_loss_clip": 0.01178017,
      "auxiliary_loss_mlp": 0.01028395,
      "balance_loss_clip": 1.01514995,
      "balance_loss_mlp": 1.02087915,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 1.6070859209395096,
      "language_loss": 0.74687082,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.76893497,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.6408848762512207
    },
    {
      "auxiliary_loss_clip": 0.01171574,
      "auxiliary_loss_mlp": 0.01029985,
      "balance_loss_clip": 0.97391367,
      "balance_loss_mlp": 1.0229038,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 1.5621328708745625,
      "language_loss": 0.72900462,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.75102019,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 2.684293270111084
    },
    {
      "auxiliary_loss_clip": 0.0116862,
      "auxiliary_loss_mlp": 0.01026432,
      "balance_loss_clip": 1.01206017,
      "balance_loss_mlp": 1.01955366,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 1.5483983625063358,
      "language_loss": 0.79835415,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.82030475,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 3.6522231101989746
    },
    {
      "auxiliary_loss_clip": 0.011737,
      "auxiliary_loss_mlp": 0.01025261,
      "balance_loss_clip": 1.05132079,
      "balance_loss_mlp": 1.01821864,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 1.502864553199058,
      "language_loss": 0.7937004,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81569004,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.659268379211426
    },
    {
      "auxiliary_loss_clip": 0.01162239,
      "auxiliary_loss_mlp": 0.01029843,
      "balance_loss_clip": 0.97124076,
      "balance_loss_mlp": 1.02271104,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 1.7790990137607678,
      "language_loss": 0.77096182,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79288268,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.6967458724975586
    },
    {
      "auxiliary_loss_clip": 0.01182507,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 0.86065423,
      "balance_loss_mlp": 1.01733756,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 2.6292619736278007,
      "language_loss": 0.81140441,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.8334806,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.759833574295044
    },
    {
      "auxiliary_loss_clip": 0.01170249,
      "auxiliary_loss_mlp": 0.0102971,
      "balance_loss_clip": 1.01189804,
      "balance_loss_mlp": 1.02271819,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 3.508136286963905,
      "language_loss": 0.67860591,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70060551,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.5742692947387695
    },
    {
      "auxiliary_loss_clip": 0.01168611,
      "auxiliary_loss_mlp": 0.01023049,
      "balance_loss_clip": 1.01374149,
      "balance_loss_mlp": 1.01618266,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.3981228932831606,
      "language_loss": 0.66234452,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.68426114,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 3.6154134273529053
    },
    {
      "auxiliary_loss_clip": 0.01166954,
      "auxiliary_loss_mlp": 0.01123084,
      "balance_loss_clip": 0.97250551,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 1.7750805697661352,
      "language_loss": 0.70218241,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.72508276,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.8574776649475098
    },
    {
      "auxiliary_loss_clip": 0.01172028,
      "auxiliary_loss_mlp": 0.01023611,
      "balance_loss_clip": 1.01161623,
      "balance_loss_mlp": 1.01655376,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 1.7782667467791837,
      "language_loss": 0.81191349,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.83386993,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.605346918106079
    },
    {
      "auxiliary_loss_clip": 0.01172854,
      "auxiliary_loss_mlp": 0.01025789,
      "balance_loss_clip": 0.93520916,
      "balance_loss_mlp": 1.01865125,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 1.8121144735561145,
      "language_loss": 0.78526461,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80725104,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 3.5895674228668213
    },
    {
      "auxiliary_loss_clip": 0.01171404,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 0.97552991,
      "balance_loss_mlp": 1.0178113,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 1.440086264705817,
      "language_loss": 0.77943492,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.8013984,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 3.506565570831299
    },
    {
      "auxiliary_loss_clip": 0.0117405,
      "auxiliary_loss_mlp": 0.01026709,
      "balance_loss_clip": 0.97371483,
      "balance_loss_mlp": 1.01919913,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 2.1235468927704595,
      "language_loss": 0.7769109,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.7989186,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.6546237468719482
    },
    {
      "auxiliary_loss_clip": 0.01170921,
      "auxiliary_loss_mlp": 0.01023928,
      "balance_loss_clip": 1.01072598,
      "balance_loss_mlp": 1.01676941,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 2.019343525534334,
      "language_loss": 0.72531438,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.7472629,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.745758295059204
    },
    {
      "auxiliary_loss_clip": 0.01176998,
      "auxiliary_loss_mlp": 0.0102831,
      "balance_loss_clip": 1.01063073,
      "balance_loss_mlp": 1.02034116,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 2.3958879398506654,
      "language_loss": 0.88317877,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90523189,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.5952954292297363
    },
    {
      "auxiliary_loss_clip": 0.0117649,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 0.9355222,
      "balance_loss_mlp": 1.02038789,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.0023443060021044,
      "language_loss": 0.67871058,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70075023,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.8327181339263916
    },
    {
      "auxiliary_loss_clip": 0.01172424,
      "auxiliary_loss_mlp": 0.01024058,
      "balance_loss_clip": 1.05086136,
      "balance_loss_mlp": 1.01724267,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 1.8986486034791459,
      "language_loss": 0.78988826,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.81185305,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 2.605787992477417
    },
    {
      "auxiliary_loss_clip": 0.01168623,
      "auxiliary_loss_mlp": 0.01020577,
      "balance_loss_clip": 0.97334981,
      "balance_loss_mlp": 1.01350832,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 1.82954556381647,
      "language_loss": 0.74795783,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.76984984,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.6954550743103027
    },
    {
      "auxiliary_loss_clip": 0.01168523,
      "auxiliary_loss_mlp": 0.01026566,
      "balance_loss_clip": 0.97158241,
      "balance_loss_mlp": 1.01955938,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 1.5716964681699417,
      "language_loss": 0.74921167,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77116257,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.703399181365967
    },
    {
      "auxiliary_loss_clip": 0.01160901,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 0.9690907,
      "balance_loss_mlp": 1.01614523,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 2.12672753220257,
      "language_loss": 0.73301136,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75486422,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 2.631713390350342
    },
    {
      "auxiliary_loss_clip": 0.0117068,
      "auxiliary_loss_mlp": 0.010233,
      "balance_loss_clip": 1.01124668,
      "balance_loss_mlp": 1.01614439,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 1.9340073981643309,
      "language_loss": 0.81291336,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83485317,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 2.659597396850586
    },
    {
      "auxiliary_loss_clip": 0.01179235,
      "auxiliary_loss_mlp": 0.01023989,
      "balance_loss_clip": 0.9745875,
      "balance_loss_mlp": 1.01628208,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 1.8350967888746383,
      "language_loss": 0.6177572,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.63978946,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 2.7843053340911865
    },
    {
      "auxiliary_loss_clip": 0.01169747,
      "auxiliary_loss_mlp": 0.01027901,
      "balance_loss_clip": 1.01148713,
      "balance_loss_mlp": 1.02055752,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 2.317093835013644,
      "language_loss": 0.88304651,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90502298,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 2.664926528930664
    },
    {
      "auxiliary_loss_clip": 0.01167988,
      "auxiliary_loss_mlp": 0.01023677,
      "balance_loss_clip": 1.01151061,
      "balance_loss_mlp": 1.0167625,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 1.816573892279501,
      "language_loss": 0.67825067,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.70016724,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.717822790145874
    },
    {
      "auxiliary_loss_clip": 0.01168934,
      "auxiliary_loss_mlp": 0.01025294,
      "balance_loss_clip": 0.8997286,
      "balance_loss_mlp": 1.01874948,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.569524277471846,
      "language_loss": 0.8359881,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.85793036,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.7401955127716064
    },
    {
      "auxiliary_loss_clip": 0.01169825,
      "auxiliary_loss_mlp": 0.01023091,
      "balance_loss_clip": 1.01337373,
      "balance_loss_mlp": 1.01598012,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 1.9229978814030595,
      "language_loss": 0.8608613,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.88279045,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 2.6726441383361816
    },
    {
      "auxiliary_loss_clip": 0.01174099,
      "auxiliary_loss_mlp": 0.01028269,
      "balance_loss_clip": 0.97337019,
      "balance_loss_mlp": 1.02105391,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 1.7241042568338305,
      "language_loss": 0.69643652,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.7184602,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.7362890243530273
    },
    {
      "auxiliary_loss_clip": 0.01171164,
      "auxiliary_loss_mlp": 0.01026009,
      "balance_loss_clip": 1.04985285,
      "balance_loss_mlp": 1.01902294,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.6049557520150999,
      "language_loss": 0.88109803,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.90306973,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 2.5907607078552246
    },
    {
      "auxiliary_loss_clip": 0.01181301,
      "auxiliary_loss_mlp": 0.01031634,
      "balance_loss_clip": 1.0542407,
      "balance_loss_mlp": 1.02371836,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 2.220173552222601,
      "language_loss": 0.86673439,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.8888638,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 3.525029182434082
    },
    {
      "auxiliary_loss_clip": 0.01175478,
      "auxiliary_loss_mlp": 0.01024146,
      "balance_loss_clip": 1.01642394,
      "balance_loss_mlp": 1.01756597,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.086129781153764,
      "language_loss": 0.69578373,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71777999,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 2.661118507385254
    },
    {
      "auxiliary_loss_clip": 0.01161551,
      "auxiliary_loss_mlp": 0.01028804,
      "balance_loss_clip": 0.96917093,
      "balance_loss_mlp": 1.02126122,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 1.409446318427351,
      "language_loss": 0.73020393,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.7521075,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.7532482147216797
    },
    {
      "auxiliary_loss_clip": 0.0117669,
      "auxiliary_loss_mlp": 0.01028021,
      "balance_loss_clip": 1.01562858,
      "balance_loss_mlp": 1.02037358,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.7092310513900153,
      "language_loss": 0.81843221,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84047937,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.6649169921875
    },
    {
      "auxiliary_loss_clip": 0.01179117,
      "auxiliary_loss_mlp": 0.01031411,
      "balance_loss_clip": 1.05392373,
      "balance_loss_mlp": 1.02372241,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.292507607394243,
      "language_loss": 0.87802529,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.90013057,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.562539577484131
    },
    {
      "auxiliary_loss_clip": 0.01085388,
      "auxiliary_loss_mlp": 0.01009162,
      "balance_loss_clip": 0.83109885,
      "balance_loss_mlp": 1.00736189,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9933767824581061,
      "language_loss": 0.65178984,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.67273533,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 4.034788131713867
    },
    {
      "auxiliary_loss_clip": 0.01181019,
      "auxiliary_loss_mlp": 0.01032333,
      "balance_loss_clip": 1.01459599,
      "balance_loss_mlp": 1.02460766,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 2.083545256088693,
      "language_loss": 0.63721377,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.6593473,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.601360321044922
    },
    {
      "auxiliary_loss_clip": 0.01176046,
      "auxiliary_loss_mlp": 0.01025415,
      "balance_loss_clip": 1.05230951,
      "balance_loss_mlp": 1.01720142,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 1.7401298886476373,
      "language_loss": 0.80298036,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.82499492,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.634106159210205
    },
    {
      "auxiliary_loss_clip": 0.01172049,
      "auxiliary_loss_mlp": 0.01026192,
      "balance_loss_clip": 1.013129,
      "balance_loss_mlp": 1.01849055,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 1.8286966539015057,
      "language_loss": 0.79775894,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.81974137,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 3.5606882572174072
    },
    {
      "auxiliary_loss_clip": 0.01172613,
      "auxiliary_loss_mlp": 0.01025666,
      "balance_loss_clip": 1.01374269,
      "balance_loss_mlp": 1.01845145,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 2.047197241299626,
      "language_loss": 0.77404892,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79603171,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 3.5945703983306885
    },
    {
      "auxiliary_loss_clip": 0.01177887,
      "auxiliary_loss_mlp": 0.01029519,
      "balance_loss_clip": 1.05187535,
      "balance_loss_mlp": 1.0217824,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 1.8478677857255037,
      "language_loss": 0.78800416,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.81007826,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 2.5828683376312256
    },
    {
      "auxiliary_loss_clip": 0.01073512,
      "auxiliary_loss_mlp": 0.010091,
      "balance_loss_clip": 1.01784801,
      "balance_loss_mlp": 1.00747836,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8902650920891464,
      "language_loss": 0.64244139,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66326749,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 2.983236074447632
    },
    {
      "auxiliary_loss_clip": 0.01171073,
      "auxiliary_loss_mlp": 0.01029029,
      "balance_loss_clip": 0.97626692,
      "balance_loss_mlp": 1.02196634,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.5757426500061504,
      "language_loss": 0.76572561,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.78772664,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.718829393386841
    },
    {
      "auxiliary_loss_clip": 0.0116584,
      "auxiliary_loss_mlp": 0.01024624,
      "balance_loss_clip": 0.93571508,
      "balance_loss_mlp": 1.01719689,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 1.7299396597179098,
      "language_loss": 0.72172272,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.74362731,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.798288583755493
    },
    {
      "auxiliary_loss_clip": 0.01075185,
      "auxiliary_loss_mlp": 0.01005236,
      "balance_loss_clip": 0.9421227,
      "balance_loss_mlp": 1.00360286,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1225930247946625,
      "language_loss": 0.55732775,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57813191,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.307067394256592
    },
    {
      "auxiliary_loss_clip": 0.01169718,
      "auxiliary_loss_mlp": 0.01025663,
      "balance_loss_clip": 0.89579856,
      "balance_loss_mlp": 1.0181706,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.6758946545306375,
      "language_loss": 0.80338699,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.82534087,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.8051931858062744
    },
    {
      "auxiliary_loss_clip": 0.01176527,
      "auxiliary_loss_mlp": 0.01027696,
      "balance_loss_clip": 0.93579972,
      "balance_loss_mlp": 1.01970279,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 1.8379666337410383,
      "language_loss": 0.74288732,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.76492959,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.744908571243286
    },
    {
      "auxiliary_loss_clip": 0.01146706,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 0.85303879,
      "balance_loss_mlp": 1.01939106,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.5880807054102906,
      "language_loss": 0.79091311,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81264764,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.8599061965942383
    },
    {
      "auxiliary_loss_clip": 0.01173506,
      "auxiliary_loss_mlp": 0.01122255,
      "balance_loss_clip": 0.93539482,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 1.876321061920366,
      "language_loss": 0.82823694,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.85119456,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 2.7250192165374756
    },
    {
      "auxiliary_loss_clip": 0.01084604,
      "auxiliary_loss_mlp": 0.01003722,
      "balance_loss_clip": 0.86769861,
      "balance_loss_mlp": 1.00189829,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.8029360249755917,
      "language_loss": 0.56085348,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.58173674,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.2937159538269043
    },
    {
      "auxiliary_loss_clip": 0.01167306,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 0.97233474,
      "balance_loss_mlp": 1.024194,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.03618860181037,
      "language_loss": 0.71395314,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.73594582,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 2.7232258319854736
    },
    {
      "auxiliary_loss_clip": 0.0117669,
      "auxiliary_loss_mlp": 0.01027156,
      "balance_loss_clip": 1.01466954,
      "balance_loss_mlp": 1.01966417,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.0639060229230366,
      "language_loss": 0.83494574,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85698414,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 2.70143985748291
    },
    {
      "auxiliary_loss_clip": 0.01172972,
      "auxiliary_loss_mlp": 0.01022628,
      "balance_loss_clip": 1.05247962,
      "balance_loss_mlp": 1.01521945,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 2.2516368609466055,
      "language_loss": 0.71783054,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.73978651,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.710498809814453
    },
    {
      "auxiliary_loss_clip": 0.01170084,
      "auxiliary_loss_mlp": 0.0102624,
      "balance_loss_clip": 0.93475616,
      "balance_loss_mlp": 1.01839614,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 1.809512110485991,
      "language_loss": 0.80969959,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.83166283,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.7812342643737793
    },
    {
      "auxiliary_loss_clip": 0.01169907,
      "auxiliary_loss_mlp": 0.01031883,
      "balance_loss_clip": 0.97370172,
      "balance_loss_mlp": 1.02418804,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.6205969511973035,
      "language_loss": 0.806171,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.8281889,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 2.702301502227783
    },
    {
      "auxiliary_loss_clip": 0.0116267,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 0.97231132,
      "balance_loss_mlp": 1.01644063,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 1.9998250328421714,
      "language_loss": 0.69867969,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.72053742,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 2.6804587841033936
    },
    {
      "auxiliary_loss_clip": 0.01166362,
      "auxiliary_loss_mlp": 0.01025192,
      "balance_loss_clip": 0.97409844,
      "balance_loss_mlp": 1.01756835,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.474628781312841,
      "language_loss": 0.66394126,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68585682,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 3.624894380569458
    },
    {
      "auxiliary_loss_clip": 0.01176449,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 1.0516417,
      "balance_loss_mlp": 1.0207175,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 2.2535648546447917,
      "language_loss": 0.78582656,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.80787528,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.6729607582092285
    },
    {
      "auxiliary_loss_clip": 0.01175511,
      "auxiliary_loss_mlp": 0.01122966,
      "balance_loss_clip": 0.89909482,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 1.9081515354110243,
      "language_loss": 0.72188669,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.7448715,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.73886775970459
    },
    {
      "auxiliary_loss_clip": 0.01184326,
      "auxiliary_loss_mlp": 0.01027166,
      "balance_loss_clip": 0.89897752,
      "balance_loss_mlp": 1.0195905,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 1.8056677649190256,
      "language_loss": 0.78512084,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.80723572,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.7831320762634277
    },
    {
      "auxiliary_loss_clip": 0.01177462,
      "auxiliary_loss_mlp": 0.01025255,
      "balance_loss_clip": 0.93719304,
      "balance_loss_mlp": 1.01727378,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 2.072730230869904,
      "language_loss": 0.75918233,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78120947,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.7938811779022217
    },
    {
      "auxiliary_loss_clip": 0.01178981,
      "auxiliary_loss_mlp": 0.01036926,
      "balance_loss_clip": 1.01581597,
      "balance_loss_mlp": 1.02922761,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.0882048441616945,
      "language_loss": 0.84756225,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.86972129,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.6713125705718994
    },
    {
      "auxiliary_loss_clip": 0.01167564,
      "auxiliary_loss_mlp": 0.01029105,
      "balance_loss_clip": 0.96907961,
      "balance_loss_mlp": 1.02173841,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 1.7914712331244955,
      "language_loss": 0.78697288,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.80893958,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 3.6498873233795166
    },
    {
      "auxiliary_loss_clip": 0.01176202,
      "auxiliary_loss_mlp": 0.01024886,
      "balance_loss_clip": 1.01299024,
      "balance_loss_mlp": 1.01746488,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.5884313555564074,
      "language_loss": 0.77926266,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80127358,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.621384620666504
    },
    {
      "auxiliary_loss_clip": 0.01172602,
      "auxiliary_loss_mlp": 0.0102575,
      "balance_loss_clip": 1.01178348,
      "balance_loss_mlp": 1.01821554,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 1.9295770472524352,
      "language_loss": 0.86291856,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.88490206,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 3.5614209175109863
    },
    {
      "auxiliary_loss_clip": 0.01175527,
      "auxiliary_loss_mlp": 0.01029799,
      "balance_loss_clip": 0.97610915,
      "balance_loss_mlp": 1.0215795,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 2.0791800313582764,
      "language_loss": 0.66471756,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.6867708,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 3.8691866397857666
    },
    {
      "auxiliary_loss_clip": 0.0116643,
      "auxiliary_loss_mlp": 0.01031345,
      "balance_loss_clip": 0.85754514,
      "balance_loss_mlp": 1.02313781,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 1.9146097553962744,
      "language_loss": 0.79238784,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81436557,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.9037067890167236
    },
    {
      "auxiliary_loss_clip": 0.01166641,
      "auxiliary_loss_mlp": 0.01026556,
      "balance_loss_clip": 0.97176707,
      "balance_loss_mlp": 1.01869106,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 1.9349151490577459,
      "language_loss": 0.71079868,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73273063,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.761176347732544
    },
    {
      "auxiliary_loss_clip": 0.0117534,
      "auxiliary_loss_mlp": 0.01027084,
      "balance_loss_clip": 0.97494471,
      "balance_loss_mlp": 1.01954961,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 1.7547523856337945,
      "language_loss": 0.57266665,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.59469086,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.631535530090332
    },
    {
      "auxiliary_loss_clip": 0.01178951,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 0.97942406,
      "balance_loss_mlp": 1.01910305,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.5085379955896852,
      "language_loss": 0.77148753,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.7935406,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.716266393661499
    },
    {
      "auxiliary_loss_clip": 0.01176092,
      "auxiliary_loss_mlp": 0.01025722,
      "balance_loss_clip": 1.05219722,
      "balance_loss_mlp": 1.01787198,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 1.762633365566265,
      "language_loss": 0.67221546,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.6942336,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 2.630439043045044
    },
    {
      "auxiliary_loss_clip": 0.01172509,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.05004263,
      "balance_loss_mlp": 1.01839149,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 2.073421471144809,
      "language_loss": 0.80644011,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82842088,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.6046838760375977
    },
    {
      "auxiliary_loss_clip": 0.01176537,
      "auxiliary_loss_mlp": 0.01122207,
      "balance_loss_clip": 1.01443326,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.4774184573057711,
      "language_loss": 0.8421979,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86518526,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.65144681930542
    },
    {
      "auxiliary_loss_clip": 0.01173549,
      "auxiliary_loss_mlp": 0.01022032,
      "balance_loss_clip": 1.01601171,
      "balance_loss_mlp": 1.01502287,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 1.852174967171871,
      "language_loss": 0.8011688,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.82312459,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.687023639678955
    },
    {
      "auxiliary_loss_clip": 0.01075942,
      "auxiliary_loss_mlp": 0.01000534,
      "balance_loss_clip": 0.93984044,
      "balance_loss_mlp": 0.99891245,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.7290814745416798,
      "language_loss": 0.5663296,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58709431,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.4053585529327393
    },
    {
      "auxiliary_loss_clip": 0.01172238,
      "auxiliary_loss_mlp": 0.01028201,
      "balance_loss_clip": 1.01243865,
      "balance_loss_mlp": 1.02074409,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 2.767487762358123,
      "language_loss": 0.76522386,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.78722829,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 2.620994806289673
    },
    {
      "auxiliary_loss_clip": 0.01172826,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.04999471,
      "balance_loss_mlp": 1.02210045,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.5567784278718602,
      "language_loss": 0.79949558,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82151985,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 2.6511850357055664
    },
    {
      "auxiliary_loss_clip": 0.01170152,
      "auxiliary_loss_mlp": 0.01027944,
      "balance_loss_clip": 0.97638935,
      "balance_loss_mlp": 1.02078271,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 1.740984618685435,
      "language_loss": 0.79469824,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.81667924,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 2.682520866394043
    },
    {
      "auxiliary_loss_clip": 0.01174321,
      "auxiliary_loss_mlp": 0.01030505,
      "balance_loss_clip": 1.01178598,
      "balance_loss_mlp": 1.02287829,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 1.8179718117668526,
      "language_loss": 0.76888973,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.79093802,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 2.643958806991577
    },
    {
      "auxiliary_loss_clip": 0.01169419,
      "auxiliary_loss_mlp": 0.01021135,
      "balance_loss_clip": 0.97073168,
      "balance_loss_mlp": 1.0142777,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 1.7139485741150702,
      "language_loss": 0.80732989,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.82923543,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.727644920349121
    },
    {
      "auxiliary_loss_clip": 0.01167296,
      "auxiliary_loss_mlp": 0.01032558,
      "balance_loss_clip": 0.97171962,
      "balance_loss_mlp": 1.02536952,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 1.6588000406296086,
      "language_loss": 0.89452255,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.91652107,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 2.6940391063690186
    },
    {
      "auxiliary_loss_clip": 0.01168886,
      "auxiliary_loss_mlp": 0.01028505,
      "balance_loss_clip": 0.89283121,
      "balance_loss_mlp": 1.02164412,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 1.8978950731376292,
      "language_loss": 0.72449327,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.74646717,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 2.9610955715179443
    },
    {
      "auxiliary_loss_clip": 0.01175172,
      "auxiliary_loss_mlp": 0.01027151,
      "balance_loss_clip": 1.05045068,
      "balance_loss_mlp": 1.0195756,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.8217146681770207,
      "language_loss": 0.69352794,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71555114,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 3.6702966690063477
    },
    {
      "auxiliary_loss_clip": 0.01179266,
      "auxiliary_loss_mlp": 0.01023033,
      "balance_loss_clip": 1.01416528,
      "balance_loss_mlp": 1.01540351,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 2.0920998655393963,
      "language_loss": 0.69307899,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71510196,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 2.5866050720214844
    },
    {
      "auxiliary_loss_clip": 0.01171526,
      "auxiliary_loss_mlp": 0.01122534,
      "balance_loss_clip": 1.0496515,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 2.235767540698618,
      "language_loss": 0.84494293,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.86788356,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 2.5809361934661865
    },
    {
      "auxiliary_loss_clip": 0.01081738,
      "auxiliary_loss_mlp": 0.01001089,
      "balance_loss_clip": 0.90299714,
      "balance_loss_mlp": 0.99940825,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7411165497851017,
      "language_loss": 0.56996459,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.59079283,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.143397092819214
    },
    {
      "auxiliary_loss_clip": 0.01176346,
      "auxiliary_loss_mlp": 0.0112268,
      "balance_loss_clip": 1.01626468,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 1.886705544179361,
      "language_loss": 0.76176947,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.78475976,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.621286392211914
    },
    {
      "auxiliary_loss_clip": 0.01170578,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 0.97199368,
      "balance_loss_mlp": 1.02174973,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.200575000912749,
      "language_loss": 0.77268505,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.79468143,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.7102293968200684
    },
    {
      "auxiliary_loss_clip": 0.01172481,
      "auxiliary_loss_mlp": 0.01027775,
      "balance_loss_clip": 0.97447199,
      "balance_loss_mlp": 1.02076828,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.674126379568804,
      "language_loss": 0.82855105,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.85055357,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 3.6590065956115723
    },
    {
      "auxiliary_loss_clip": 0.0117526,
      "auxiliary_loss_mlp": 0.01025942,
      "balance_loss_clip": 0.97298682,
      "balance_loss_mlp": 1.01869154,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 1.6541726339979037,
      "language_loss": 0.81678057,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83879262,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.6811037063598633
    },
    {
      "auxiliary_loss_clip": 0.01172533,
      "auxiliary_loss_mlp": 0.01027023,
      "balance_loss_clip": 0.85918546,
      "balance_loss_mlp": 1.02018344,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.5894906218066958,
      "language_loss": 0.83233786,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.8543334,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 3.70139217376709
    },
    {
      "auxiliary_loss_clip": 0.01175921,
      "auxiliary_loss_mlp": 0.01122081,
      "balance_loss_clip": 1.01236725,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 1.9108505096982882,
      "language_loss": 0.66458142,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.68756145,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 3.6610803604125977
    },
    {
      "auxiliary_loss_clip": 0.01180916,
      "auxiliary_loss_mlp": 0.01032231,
      "balance_loss_clip": 0.97446507,
      "balance_loss_mlp": 1.02490592,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 2.0944370307982605,
      "language_loss": 0.80823696,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.83036846,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.6877400875091553
    },
    {
      "auxiliary_loss_clip": 0.01176421,
      "auxiliary_loss_mlp": 0.01024755,
      "balance_loss_clip": 1.01212788,
      "balance_loss_mlp": 1.01769781,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 2.0195893711407926,
      "language_loss": 0.72591841,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74793017,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 2.65054988861084
    },
    {
      "auxiliary_loss_clip": 0.01163094,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 0.937841,
      "balance_loss_mlp": 1.02126312,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 1.919872407201989,
      "language_loss": 0.80773991,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.82965946,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.6977415084838867
    },
    {
      "auxiliary_loss_clip": 0.01173507,
      "auxiliary_loss_mlp": 0.01025221,
      "balance_loss_clip": 1.05054379,
      "balance_loss_mlp": 1.01800311,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 2.2513723489785775,
      "language_loss": 0.82194388,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.8439312,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.650275230407715
    },
    {
      "auxiliary_loss_clip": 0.01169735,
      "auxiliary_loss_mlp": 0.01022135,
      "balance_loss_clip": 1.01164484,
      "balance_loss_mlp": 1.01528955,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 1.7851617082728015,
      "language_loss": 0.79827917,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.82019788,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 2.624985456466675
    },
    {
      "auxiliary_loss_clip": 0.01174733,
      "auxiliary_loss_mlp": 0.01019725,
      "balance_loss_clip": 1.01247573,
      "balance_loss_mlp": 1.01320159,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 2.4745377288321597,
      "language_loss": 0.70492554,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.72687012,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.668116807937622
    },
    {
      "auxiliary_loss_clip": 0.01176349,
      "auxiliary_loss_mlp": 0.01029946,
      "balance_loss_clip": 1.01149702,
      "balance_loss_mlp": 1.02198565,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 2.97760730230223,
      "language_loss": 0.79512972,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.81719267,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.663921594619751
    },
    {
      "auxiliary_loss_clip": 0.0117017,
      "auxiliary_loss_mlp": 0.01027487,
      "balance_loss_clip": 1.0098455,
      "balance_loss_mlp": 1.02034044,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 1.7908560047202573,
      "language_loss": 0.8525784,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.87455499,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.7054617404937744
    },
    {
      "auxiliary_loss_clip": 0.01164359,
      "auxiliary_loss_mlp": 0.01027583,
      "balance_loss_clip": 0.93505383,
      "balance_loss_mlp": 1.02009034,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 1.473502040998763,
      "language_loss": 0.66633046,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68824989,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.69323992729187
    },
    {
      "auxiliary_loss_clip": 0.01169476,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 0.97022712,
      "balance_loss_mlp": 1.01753807,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.3888761105132317,
      "language_loss": 0.77443475,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.79637313,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 2.7846755981445312
    },
    {
      "auxiliary_loss_clip": 0.01168472,
      "auxiliary_loss_mlp": 0.01026833,
      "balance_loss_clip": 0.97265017,
      "balance_loss_mlp": 1.019382,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 1.8004729284966274,
      "language_loss": 0.57419157,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.59614468,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.7697336673736572
    },
    {
      "auxiliary_loss_clip": 0.01167875,
      "auxiliary_loss_mlp": 0.01022858,
      "balance_loss_clip": 0.9735707,
      "balance_loss_mlp": 1.01583362,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 3.5321157212198715,
      "language_loss": 0.82573199,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.84763932,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 2.7610888481140137
    },
    {
      "auxiliary_loss_clip": 0.01168234,
      "auxiliary_loss_mlp": 0.01025118,
      "balance_loss_clip": 0.97173786,
      "balance_loss_mlp": 1.01772738,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.0998529844703877,
      "language_loss": 0.78325427,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80518782,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.7562217712402344
    },
    {
      "auxiliary_loss_clip": 0.01159616,
      "auxiliary_loss_mlp": 0.01026251,
      "balance_loss_clip": 0.97157931,
      "balance_loss_mlp": 1.01917005,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 1.8567338691241775,
      "language_loss": 0.74348867,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76534736,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 2.688044786453247
    },
    {
      "auxiliary_loss_clip": 0.01169991,
      "auxiliary_loss_mlp": 0.01025045,
      "balance_loss_clip": 1.01148343,
      "balance_loss_mlp": 1.01739717,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 2.7635843139706844,
      "language_loss": 0.89654171,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.91849214,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 2.645669937133789
    },
    {
      "auxiliary_loss_clip": 0.01166003,
      "auxiliary_loss_mlp": 0.01032399,
      "balance_loss_clip": 0.97222477,
      "balance_loss_mlp": 1.023947,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 2.1055871669481814,
      "language_loss": 0.63553512,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.65751922,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 2.704340696334839
    },
    {
      "auxiliary_loss_clip": 0.01174619,
      "auxiliary_loss_mlp": 0.01032311,
      "balance_loss_clip": 0.97597891,
      "balance_loss_mlp": 1.02463436,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 1.6975241086763877,
      "language_loss": 0.84439045,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.86645973,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 3.632917642593384
    },
    {
      "auxiliary_loss_clip": 0.01178847,
      "auxiliary_loss_mlp": 0.01122525,
      "balance_loss_clip": 1.0142746,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 1.7420837986804534,
      "language_loss": 0.76289731,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78591108,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.6323928833007812
    },
    {
      "auxiliary_loss_clip": 0.01176752,
      "auxiliary_loss_mlp": 0.01122672,
      "balance_loss_clip": 1.01570344,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.5854140698275405,
      "language_loss": 0.8054769,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.82847118,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 2.6671388149261475
    },
    {
      "auxiliary_loss_clip": 0.01171142,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.01259851,
      "balance_loss_mlp": 1.01994503,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.6137245890507805,
      "language_loss": 0.72783744,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.74982554,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.674422025680542
    },
    {
      "auxiliary_loss_clip": 0.01167212,
      "auxiliary_loss_mlp": 0.0112241,
      "balance_loss_clip": 0.85753787,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 1.9673771412754826,
      "language_loss": 0.73426574,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.75716197,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.676201343536377
    },
    {
      "auxiliary_loss_clip": 0.01172495,
      "auxiliary_loss_mlp": 0.01122911,
      "balance_loss_clip": 1.05022478,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 1.8821935451429754,
      "language_loss": 0.67615056,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.69910455,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.7006118297576904
    },
    {
      "auxiliary_loss_clip": 0.01166062,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 0.8956784,
      "balance_loss_mlp": 1.01962101,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 1.7085380168535442,
      "language_loss": 0.8778528,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.89977956,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 3.636960506439209
    },
    {
      "auxiliary_loss_clip": 0.01069743,
      "auxiliary_loss_mlp": 0.01116659,
      "balance_loss_clip": 1.01477599,
      "balance_loss_mlp": 0.0,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7848478805896684,
      "language_loss": 0.61825937,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.64012343,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.301440954208374
    },
    {
      "auxiliary_loss_clip": 0.01166926,
      "auxiliary_loss_mlp": 0.01032385,
      "balance_loss_clip": 0.9349696,
      "balance_loss_mlp": 1.02437735,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 2.5025324558968807,
      "language_loss": 0.7839421,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80593526,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 3.5244929790496826
    },
    {
      "auxiliary_loss_clip": 0.0117064,
      "auxiliary_loss_mlp": 0.01028343,
      "balance_loss_clip": 0.97093517,
      "balance_loss_mlp": 1.02033758,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 1.9924182292173762,
      "language_loss": 0.73880905,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.76079887,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 3.6357271671295166
    },
    {
      "auxiliary_loss_clip": 0.01171885,
      "auxiliary_loss_mlp": 0.01022594,
      "balance_loss_clip": 1.05081391,
      "balance_loss_mlp": 1.01585019,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 1.7015265233048205,
      "language_loss": 0.74250734,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76445222,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.6948301792144775
    },
    {
      "auxiliary_loss_clip": 0.01171792,
      "auxiliary_loss_mlp": 0.01023467,
      "balance_loss_clip": 1.01030231,
      "balance_loss_mlp": 1.01606452,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 1.767845072960063,
      "language_loss": 0.67995071,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.70190322,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.6700191497802734
    },
    {
      "auxiliary_loss_clip": 0.01167028,
      "auxiliary_loss_mlp": 0.01036775,
      "balance_loss_clip": 0.93623108,
      "balance_loss_mlp": 1.0296371,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 1.4969028019130797,
      "language_loss": 0.73391008,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75594813,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.8364691734313965
    },
    {
      "auxiliary_loss_clip": 0.01173254,
      "auxiliary_loss_mlp": 0.01027372,
      "balance_loss_clip": 1.01270652,
      "balance_loss_mlp": 1.01995182,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 1.9366642761244772,
      "language_loss": 0.79192799,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81393427,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.6875314712524414
    },
    {
      "auxiliary_loss_clip": 0.01074545,
      "auxiliary_loss_mlp": 0.01003275,
      "balance_loss_clip": 0.97780085,
      "balance_loss_mlp": 1.00161827,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.7751151997092094,
      "language_loss": 0.59622079,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61699903,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.2700698375701904
    },
    {
      "auxiliary_loss_clip": 0.01173067,
      "auxiliary_loss_mlp": 0.01027744,
      "balance_loss_clip": 1.01290298,
      "balance_loss_mlp": 1.01997399,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 1.978546614901157,
      "language_loss": 0.7171275,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.73913562,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.661505937576294
    },
    {
      "auxiliary_loss_clip": 0.01172155,
      "auxiliary_loss_mlp": 0.01022086,
      "balance_loss_clip": 0.97266793,
      "balance_loss_mlp": 1.01472461,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 1.7221468075319668,
      "language_loss": 0.62857008,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.65051246,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.762211561203003
    },
    {
      "auxiliary_loss_clip": 0.01173991,
      "auxiliary_loss_mlp": 0.0103337,
      "balance_loss_clip": 0.97536236,
      "balance_loss_mlp": 1.02526999,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 2.017967584475808,
      "language_loss": 0.85509092,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.87716454,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 2.7046072483062744
    },
    {
      "auxiliary_loss_clip": 0.0117723,
      "auxiliary_loss_mlp": 0.01026079,
      "balance_loss_clip": 1.0521481,
      "balance_loss_mlp": 1.01862574,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 1.8503433978194652,
      "language_loss": 0.79560262,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.81763577,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.595560073852539
    },
    {
      "auxiliary_loss_clip": 0.0116824,
      "auxiliary_loss_mlp": 0.0103206,
      "balance_loss_clip": 0.97623461,
      "balance_loss_mlp": 1.02534842,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 1.886255005004411,
      "language_loss": 0.7129885,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73499143,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.6846907138824463
    },
    {
      "auxiliary_loss_clip": 0.01181427,
      "auxiliary_loss_mlp": 0.01026268,
      "balance_loss_clip": 1.01482093,
      "balance_loss_mlp": 1.01824522,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 2.042527968632584,
      "language_loss": 0.67886126,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.70093822,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 2.8002097606658936
    },
    {
      "auxiliary_loss_clip": 0.01169629,
      "auxiliary_loss_mlp": 0.0103124,
      "balance_loss_clip": 0.97379827,
      "balance_loss_mlp": 1.02320552,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 1.545299629711732,
      "language_loss": 0.7342909,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75629956,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 2.693079710006714
    },
    {
      "auxiliary_loss_clip": 0.0116287,
      "auxiliary_loss_mlp": 0.01028061,
      "balance_loss_clip": 0.97361225,
      "balance_loss_mlp": 1.02083373,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 1.7926706498157896,
      "language_loss": 0.69537604,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.71728545,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 2.7795684337615967
    },
    {
      "auxiliary_loss_clip": 0.01170911,
      "auxiliary_loss_mlp": 0.01024191,
      "balance_loss_clip": 1.00870013,
      "balance_loss_mlp": 1.01686859,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.7401807070428719,
      "language_loss": 0.86440837,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88635939,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.663518190383911
    },
    {
      "auxiliary_loss_clip": 0.01166102,
      "auxiliary_loss_mlp": 0.01028995,
      "balance_loss_clip": 0.97224051,
      "balance_loss_mlp": 1.02191734,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 1.8398936119166531,
      "language_loss": 0.86145794,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.8834089,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 2.6708157062530518
    },
    {
      "auxiliary_loss_clip": 0.01175297,
      "auxiliary_loss_mlp": 0.01030482,
      "balance_loss_clip": 0.93875128,
      "balance_loss_mlp": 1.02269483,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 1.8082640152543152,
      "language_loss": 0.76183939,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78389716,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.6663479804992676
    },
    {
      "auxiliary_loss_clip": 0.0117078,
      "auxiliary_loss_mlp": 0.01034423,
      "balance_loss_clip": 0.93731076,
      "balance_loss_mlp": 1.0275805,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 2.020088534817219,
      "language_loss": 0.83620018,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.85825223,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 3.6031906604766846
    },
    {
      "auxiliary_loss_clip": 0.01174049,
      "auxiliary_loss_mlp": 0.0112328,
      "balance_loss_clip": 1.01416135,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.4752142108655906,
      "language_loss": 0.75484419,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.77781749,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.6706173419952393
    },
    {
      "auxiliary_loss_clip": 0.01170168,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.01349235,
      "balance_loss_mlp": 1.02027464,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 1.9640426248201777,
      "language_loss": 0.82915705,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.85112977,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 2.7085421085357666
    },
    {
      "auxiliary_loss_clip": 0.01163688,
      "auxiliary_loss_mlp": 0.01030983,
      "balance_loss_clip": 0.97152328,
      "balance_loss_mlp": 1.02371395,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 2.64396174167258,
      "language_loss": 0.7742753,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.79622197,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.692646026611328
    },
    {
      "auxiliary_loss_clip": 0.01167137,
      "auxiliary_loss_mlp": 0.0102873,
      "balance_loss_clip": 0.97156531,
      "balance_loss_mlp": 1.02184558,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 2.241347242159276,
      "language_loss": 0.80264747,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.82460618,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.797010898590088
    },
    {
      "auxiliary_loss_clip": 0.01168122,
      "auxiliary_loss_mlp": 0.01122402,
      "balance_loss_clip": 0.97199988,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 2.001622475147847,
      "language_loss": 0.66566944,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.68857473,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.7452962398529053
    },
    {
      "auxiliary_loss_clip": 0.01171401,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 1.05050397,
      "balance_loss_mlp": 1.0165112,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 2.05807261471852,
      "language_loss": 0.86203045,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.88398099,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 3.5626401901245117
    },
    {
      "auxiliary_loss_clip": 0.01172202,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.04912615,
      "balance_loss_mlp": 1.01802468,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 1.7667891035940275,
      "language_loss": 0.68361664,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70559394,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.637964963912964
    },
    {
      "auxiliary_loss_clip": 0.01172212,
      "auxiliary_loss_mlp": 0.0102866,
      "balance_loss_clip": 0.97606254,
      "balance_loss_mlp": 1.02120376,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 2.240675031982481,
      "language_loss": 0.74107134,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76308,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 3.6528568267822266
    },
    {
      "auxiliary_loss_clip": 0.0115354,
      "auxiliary_loss_mlp": 0.01026056,
      "balance_loss_clip": 0.85608947,
      "balance_loss_mlp": 1.01913285,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.6199977450527692,
      "language_loss": 0.77020895,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.79200494,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 3.618755340576172
    },
    {
      "auxiliary_loss_clip": 0.01171133,
      "auxiliary_loss_mlp": 0.01024463,
      "balance_loss_clip": 1.01026559,
      "balance_loss_mlp": 1.01684523,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.3964976700036726,
      "language_loss": 0.70685643,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.7288124,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.7206602096557617
    },
    {
      "auxiliary_loss_clip": 0.01175322,
      "auxiliary_loss_mlp": 0.01027099,
      "balance_loss_clip": 1.01234889,
      "balance_loss_mlp": 1.01970172,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 2.110011760531482,
      "language_loss": 0.87058616,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.89261031,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.6321871280670166
    },
    {
      "auxiliary_loss_clip": 0.01171682,
      "auxiliary_loss_mlp": 0.01123042,
      "balance_loss_clip": 1.01213026,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 6.669028647632212,
      "language_loss": 0.63925165,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.6621989,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.652977466583252
    },
    {
      "auxiliary_loss_clip": 0.01173183,
      "auxiliary_loss_mlp": 0.01033894,
      "balance_loss_clip": 0.93348777,
      "balance_loss_mlp": 1.02633584,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 1.8614321725895009,
      "language_loss": 0.89956957,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.9216404,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.822801351547241
    },
    {
      "auxiliary_loss_clip": 0.01171593,
      "auxiliary_loss_mlp": 0.01023335,
      "balance_loss_clip": 0.97529823,
      "balance_loss_mlp": 1.01629281,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.9807999912063095,
      "language_loss": 0.64014786,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.6620971,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.6235291957855225
    },
    {
      "auxiliary_loss_clip": 0.01164098,
      "auxiliary_loss_mlp": 0.01021601,
      "balance_loss_clip": 0.97199178,
      "balance_loss_mlp": 1.01442814,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.011165595149073,
      "language_loss": 0.70677853,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.72863549,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.7040295600891113
    },
    {
      "auxiliary_loss_clip": 0.01169341,
      "auxiliary_loss_mlp": 0.01025425,
      "balance_loss_clip": 0.9714793,
      "balance_loss_mlp": 1.0184809,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 1.7215224130824318,
      "language_loss": 0.80957627,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.8315239,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.6670219898223877
    },
    {
      "auxiliary_loss_clip": 0.01162392,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.0103538,
      "balance_loss_mlp": 1.01935697,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 1.9061012857603425,
      "language_loss": 0.72178036,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74367166,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.665391445159912
    },
    {
      "auxiliary_loss_clip": 0.01174634,
      "auxiliary_loss_mlp": 0.01026783,
      "balance_loss_clip": 0.89998084,
      "balance_loss_mlp": 1.01881373,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.6618332504922355,
      "language_loss": 0.72561365,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.74762785,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 2.750819683074951
    },
    {
      "auxiliary_loss_clip": 0.01169957,
      "auxiliary_loss_mlp": 0.01026683,
      "balance_loss_clip": 0.97393173,
      "balance_loss_mlp": 1.01962578,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.1053839284900224,
      "language_loss": 0.72794664,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.74991304,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 2.6712186336517334
    },
    {
      "auxiliary_loss_clip": 0.01171158,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04830945,
      "balance_loss_mlp": 1.01803637,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.6205421952834211,
      "language_loss": 0.80039573,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.82235277,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 2.611116647720337
    },
    {
      "auxiliary_loss_clip": 0.0117707,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 0.97494054,
      "balance_loss_mlp": 1.01953125,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 2.4552939046626525,
      "language_loss": 0.84125072,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.8632893,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 2.656154155731201
    },
    {
      "auxiliary_loss_clip": 0.01170758,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.01299357,
      "balance_loss_mlp": 1.02311659,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.9978569459669984,
      "language_loss": 0.73902637,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76103562,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 2.6134183406829834
    },
    {
      "auxiliary_loss_clip": 0.01157703,
      "auxiliary_loss_mlp": 0.01025181,
      "balance_loss_clip": 0.85634327,
      "balance_loss_mlp": 1.01819205,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 1.855379600672011,
      "language_loss": 0.76487565,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78670454,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 2.6474483013153076
    },
    {
      "auxiliary_loss_clip": 0.01175966,
      "auxiliary_loss_mlp": 0.01025294,
      "balance_loss_clip": 1.05262089,
      "balance_loss_mlp": 1.01787364,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 1.7711258762092028,
      "language_loss": 0.76404756,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78606015,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 2.5164992809295654
    },
    {
      "auxiliary_loss_clip": 0.01075964,
      "auxiliary_loss_mlp": 0.0100052,
      "balance_loss_clip": 0.90303349,
      "balance_loss_mlp": 0.99883884,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.7606024628554326,
      "language_loss": 0.56480408,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.5855689,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.2080395221710205
    },
    {
      "auxiliary_loss_clip": 0.0117576,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.05129647,
      "balance_loss_mlp": 1.01895475,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 1.9573158995543414,
      "language_loss": 0.89272833,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91474676,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 3.5111851692199707
    },
    {
      "auxiliary_loss_clip": 0.01069863,
      "auxiliary_loss_mlp": 0.01001911,
      "balance_loss_clip": 1.01516581,
      "balance_loss_mlp": 1.00030124,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 0.8264667938778889,
      "language_loss": 0.56099391,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58171165,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.1935343742370605
    },
    {
      "auxiliary_loss_clip": 0.01173857,
      "auxiliary_loss_mlp": 0.01024025,
      "balance_loss_clip": 1.01352298,
      "balance_loss_mlp": 1.01663411,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 1.7796261489002208,
      "language_loss": 0.80673665,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.82871544,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 2.6606602668762207
    },
    {
      "auxiliary_loss_clip": 0.01175481,
      "auxiliary_loss_mlp": 0.01024861,
      "balance_loss_clip": 1.01185167,
      "balance_loss_mlp": 1.01792848,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.7622741866874854,
      "language_loss": 0.7593534,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.78135681,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.6542890071868896
    },
    {
      "auxiliary_loss_clip": 0.01172336,
      "auxiliary_loss_mlp": 0.01024474,
      "balance_loss_clip": 1.05014312,
      "balance_loss_mlp": 1.01755977,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 1.5078353673139473,
      "language_loss": 0.82457471,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84654284,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.581233024597168
    },
    {
      "auxiliary_loss_clip": 0.01177836,
      "auxiliary_loss_mlp": 0.01122904,
      "balance_loss_clip": 0.97518152,
      "balance_loss_mlp": 0.0,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 1.7692129612677834,
      "language_loss": 0.70893961,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.731947,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.64294171333313
    },
    {
      "auxiliary_loss_clip": 0.01160137,
      "auxiliary_loss_mlp": 0.01027262,
      "balance_loss_clip": 0.85614228,
      "balance_loss_mlp": 1.01990724,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 1.6665374231107406,
      "language_loss": 0.75786424,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77973819,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 3.604705572128296
    },
    {
      "auxiliary_loss_clip": 0.01170858,
      "auxiliary_loss_mlp": 0.0102786,
      "balance_loss_clip": 0.97401655,
      "balance_loss_mlp": 1.02072191,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 1.6453927341934536,
      "language_loss": 0.76231122,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78429842,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.6185405254364014
    },
    {
      "auxiliary_loss_clip": 0.01165373,
      "auxiliary_loss_mlp": 0.01031411,
      "balance_loss_clip": 0.89665323,
      "balance_loss_mlp": 1.02384686,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.511572468467276,
      "language_loss": 0.72510517,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74707305,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 3.7675373554229736
    },
    {
      "auxiliary_loss_clip": 0.01173656,
      "auxiliary_loss_mlp": 0.01026064,
      "balance_loss_clip": 1.01404881,
      "balance_loss_mlp": 1.018327,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.279009500653632,
      "language_loss": 0.71765614,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.73965341,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 3.590041160583496
    },
    {
      "auxiliary_loss_clip": 0.01177093,
      "auxiliary_loss_mlp": 0.01029024,
      "balance_loss_clip": 1.05204129,
      "balance_loss_mlp": 1.02205646,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.4464099013490201,
      "language_loss": 0.78534836,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.80740952,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.8230514526367188
    },
    {
      "auxiliary_loss_clip": 0.01171491,
      "auxiliary_loss_mlp": 0.0102631,
      "balance_loss_clip": 0.97426039,
      "balance_loss_mlp": 1.01908302,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 2.834434575032191,
      "language_loss": 0.73592675,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.75790477,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.6731679439544678
    },
    {
      "auxiliary_loss_clip": 0.01162905,
      "auxiliary_loss_mlp": 0.01031238,
      "balance_loss_clip": 0.93158114,
      "balance_loss_mlp": 1.02351916,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 1.9441324720704534,
      "language_loss": 0.85185182,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.87379324,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 321.68075823783875
    },
    {
      "auxiliary_loss_clip": 0.01176228,
      "auxiliary_loss_mlp": 0.01029211,
      "balance_loss_clip": 1.01128864,
      "balance_loss_mlp": 1.02063429,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.0979119606071586,
      "language_loss": 0.70800781,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.73006219,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.704411745071411
    },
    {
      "auxiliary_loss_clip": 0.01165984,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 0.93778384,
      "balance_loss_mlp": 1.02075624,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 1.796928660662483,
      "language_loss": 0.75845385,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.78039241,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.7671751976013184
    },
    {
      "auxiliary_loss_clip": 0.01167706,
      "auxiliary_loss_mlp": 0.01027994,
      "balance_loss_clip": 0.97671103,
      "balance_loss_mlp": 1.02115774,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 1.6665760995198922,
      "language_loss": 0.77365005,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.79560703,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.757735252380371
    },
    {
      "auxiliary_loss_clip": 0.01172147,
      "auxiliary_loss_mlp": 0.010305,
      "balance_loss_clip": 1.01291728,
      "balance_loss_mlp": 1.0228529,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 1.5604417571535965,
      "language_loss": 0.8186152,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.84064162,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.7172293663024902
    },
    {
      "auxiliary_loss_clip": 0.01167781,
      "auxiliary_loss_mlp": 0.01032776,
      "balance_loss_clip": 0.97374141,
      "balance_loss_mlp": 1.025105,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 2.3524746151510794,
      "language_loss": 0.84136719,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86337262,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.6787848472595215
    },
    {
      "auxiliary_loss_clip": 0.01166531,
      "auxiliary_loss_mlp": 0.01028716,
      "balance_loss_clip": 1.012187,
      "balance_loss_mlp": 1.02136111,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 2.2563306863911414,
      "language_loss": 0.86241907,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.88437158,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.7049922943115234
    },
    {
      "auxiliary_loss_clip": 0.0117903,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.01291537,
      "balance_loss_mlp": 1.01773691,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 1.736882516438286,
      "language_loss": 0.7392087,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.761253,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 2.6522634029388428
    },
    {
      "auxiliary_loss_clip": 0.01173407,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 1.0519253,
      "balance_loss_mlp": 1.02499473,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 1.8430172236746492,
      "language_loss": 0.60120678,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62326068,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 2.5957906246185303
    },
    {
      "auxiliary_loss_clip": 0.01165414,
      "auxiliary_loss_mlp": 0.01034683,
      "balance_loss_clip": 0.9333849,
      "balance_loss_mlp": 1.02764034,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.212631110157406,
      "language_loss": 0.69169533,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.7136963,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.718151807785034
    },
    {
      "auxiliary_loss_clip": 0.0117068,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.01357484,
      "balance_loss_mlp": 1.01961839,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 1.8651789510017303,
      "language_loss": 0.6903019,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.71227813,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 2.678391933441162
    },
    {
      "auxiliary_loss_clip": 0.01176152,
      "auxiliary_loss_mlp": 0.01026343,
      "balance_loss_clip": 0.89482534,
      "balance_loss_mlp": 1.01834965,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.754389267125356,
      "language_loss": 0.74282336,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.76484835,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 2.748781204223633
    },
    {
      "auxiliary_loss_clip": 0.01173354,
      "auxiliary_loss_mlp": 0.01032076,
      "balance_loss_clip": 0.97412777,
      "balance_loss_mlp": 1.0237608,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 1.6878071813549558,
      "language_loss": 0.7317332,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75378752,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 2.671736478805542
    },
    {
      "auxiliary_loss_clip": 0.01163271,
      "auxiliary_loss_mlp": 0.01024179,
      "balance_loss_clip": 0.93151367,
      "balance_loss_mlp": 1.01733947,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 2.127298622499147,
      "language_loss": 0.62797213,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.64984661,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 2.8040719032287598
    },
    {
      "auxiliary_loss_clip": 0.01073181,
      "auxiliary_loss_mlp": 0.01012366,
      "balance_loss_clip": 0.9021076,
      "balance_loss_mlp": 1.0108937,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.7949881992888984,
      "language_loss": 0.5586074,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57946283,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 4.191433429718018
    },
    {
      "auxiliary_loss_clip": 0.01170214,
      "auxiliary_loss_mlp": 0.01034285,
      "balance_loss_clip": 0.97681576,
      "balance_loss_mlp": 1.02720976,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 2.014361084075841,
      "language_loss": 0.66780698,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.689852,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 2.6844873428344727
    },
    {
      "auxiliary_loss_clip": 0.01170541,
      "auxiliary_loss_mlp": 0.01122473,
      "balance_loss_clip": 1.05075526,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 1.656021281556466,
      "language_loss": 0.80917406,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.83210421,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.683014392852783
    },
    {
      "auxiliary_loss_clip": 0.01167111,
      "auxiliary_loss_mlp": 0.01122777,
      "balance_loss_clip": 0.97638988,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 1.8590208455059514,
      "language_loss": 0.82114869,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.84404755,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 2.6880130767822266
    },
    {
      "auxiliary_loss_clip": 0.01174883,
      "auxiliary_loss_mlp": 0.01028302,
      "balance_loss_clip": 1.01160014,
      "balance_loss_mlp": 1.02082467,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 1.6056521010592602,
      "language_loss": 0.82147932,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84351122,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.636472463607788
    },
    {
      "auxiliary_loss_clip": 0.01166198,
      "auxiliary_loss_mlp": 0.01026027,
      "balance_loss_clip": 1.01078081,
      "balance_loss_mlp": 1.01869559,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.583064274549858,
      "language_loss": 0.85995436,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.88187665,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 3.6496455669403076
    },
    {
      "auxiliary_loss_clip": 0.01170788,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.01333976,
      "balance_loss_mlp": 1.0222435,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 1.6998217575796017,
      "language_loss": 0.72765535,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.74965614,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.5956473350524902
    },
    {
      "auxiliary_loss_clip": 0.01167896,
      "auxiliary_loss_mlp": 0.01031327,
      "balance_loss_clip": 0.74541277,
      "balance_loss_mlp": 1.02435911,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 1.9861655010421113,
      "language_loss": 0.80186391,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.82385612,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 4.00597071647644
    },
    {
      "auxiliary_loss_clip": 0.01168565,
      "auxiliary_loss_mlp": 0.01026184,
      "balance_loss_clip": 0.97619456,
      "balance_loss_mlp": 1.01931119,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 3.8811942703955333,
      "language_loss": 0.77908659,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.80103409,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 4.112343788146973
    },
    {
      "auxiliary_loss_clip": 0.01170076,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 0.97460073,
      "balance_loss_mlp": 1.01985669,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.2740653326221145,
      "language_loss": 0.8053689,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.82734013,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.712754726409912
    },
    {
      "auxiliary_loss_clip": 0.01163316,
      "auxiliary_loss_mlp": 0.0102624,
      "balance_loss_clip": 1.00842643,
      "balance_loss_mlp": 1.01873541,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 1.971466593697614,
      "language_loss": 0.82277751,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84467304,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.6487503051757812
    },
    {
      "auxiliary_loss_clip": 0.01154343,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 0.855066,
      "balance_loss_mlp": 1.01708412,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 1.6573828644384965,
      "language_loss": 0.77372587,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.79551739,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 2.787829637527466
    },
    {
      "auxiliary_loss_clip": 0.01166505,
      "auxiliary_loss_mlp": 0.01024976,
      "balance_loss_clip": 0.81538922,
      "balance_loss_mlp": 1.0178895,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 2.040986518892681,
      "language_loss": 0.7989558,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.82087064,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.855949878692627
    },
    {
      "auxiliary_loss_clip": 0.01168069,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.01114678,
      "balance_loss_mlp": 1.02208686,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 2.290470542444935,
      "language_loss": 0.72464275,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74661773,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 2.6892223358154297
    },
    {
      "auxiliary_loss_clip": 0.01173858,
      "auxiliary_loss_mlp": 0.01026012,
      "balance_loss_clip": 0.93386316,
      "balance_loss_mlp": 1.01829362,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 2.265679941164304,
      "language_loss": 0.67870545,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.7007041,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.828340530395508
    },
    {
      "auxiliary_loss_clip": 0.01166187,
      "auxiliary_loss_mlp": 0.01123011,
      "balance_loss_clip": 0.97188616,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 1.7357803018255762,
      "language_loss": 0.71373785,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.73662984,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.669541597366333
    },
    {
      "auxiliary_loss_clip": 0.01080301,
      "auxiliary_loss_mlp": 0.01000131,
      "balance_loss_clip": 0.90401125,
      "balance_loss_mlp": 0.9984979,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.9021438431877629,
      "language_loss": 0.62742203,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64822632,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.123610258102417
    },
    {
      "auxiliary_loss_clip": 0.01068911,
      "auxiliary_loss_mlp": 0.01001096,
      "balance_loss_clip": 1.01381063,
      "balance_loss_mlp": 0.99943876,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8815009190903949,
      "language_loss": 0.58555567,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60625571,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.950090169906616
    },
    {
      "auxiliary_loss_clip": 0.01171711,
      "auxiliary_loss_mlp": 0.01028769,
      "balance_loss_clip": 1.00852597,
      "balance_loss_mlp": 1.02162838,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 1.9028580848209,
      "language_loss": 0.73967886,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.7616837,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.734583616256714
    },
    {
      "auxiliary_loss_clip": 0.01170937,
      "auxiliary_loss_mlp": 0.01122376,
      "balance_loss_clip": 1.04976773,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 1.6915122553409727,
      "language_loss": 0.66420937,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.68714249,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 2.836838722229004
    },
    {
      "auxiliary_loss_clip": 0.01176462,
      "auxiliary_loss_mlp": 0.0103143,
      "balance_loss_clip": 1.01273489,
      "balance_loss_mlp": 1.0239557,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 1.8913788564754954,
      "language_loss": 0.77298802,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79506695,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 2.64276123046875
    },
    {
      "auxiliary_loss_clip": 0.01164736,
      "auxiliary_loss_mlp": 0.01026418,
      "balance_loss_clip": 0.96889389,
      "balance_loss_mlp": 1.01916373,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 1.9715575311832427,
      "language_loss": 0.77766597,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.79957759,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 2.699321746826172
    },
    {
      "auxiliary_loss_clip": 0.01176474,
      "auxiliary_loss_mlp": 0.01025227,
      "balance_loss_clip": 0.93603289,
      "balance_loss_mlp": 1.01752019,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 1.9242093916438054,
      "language_loss": 0.79589856,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.81791556,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 2.677422046661377
    },
    {
      "auxiliary_loss_clip": 0.01176091,
      "auxiliary_loss_mlp": 0.01031069,
      "balance_loss_clip": 1.01100326,
      "balance_loss_mlp": 1.02355242,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 1.645405166339408,
      "language_loss": 0.72068191,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74275351,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 2.7149698734283447
    },
    {
      "auxiliary_loss_clip": 0.01171465,
      "auxiliary_loss_mlp": 0.01032689,
      "balance_loss_clip": 1.00939119,
      "balance_loss_mlp": 1.02444577,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.432828571300034,
      "language_loss": 0.69765985,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.71970141,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 2.6996657848358154
    },
    {
      "auxiliary_loss_clip": 0.01171098,
      "auxiliary_loss_mlp": 0.0102532,
      "balance_loss_clip": 1.05042434,
      "balance_loss_mlp": 1.01828063,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.1118502022580907,
      "language_loss": 0.7862289,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.80819309,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 2.6895368099212646
    },
    {
      "auxiliary_loss_clip": 0.01162509,
      "auxiliary_loss_mlp": 0.01024553,
      "balance_loss_clip": 0.9697845,
      "balance_loss_mlp": 1.01783299,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 2.7093102160503824,
      "language_loss": 0.75189137,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77376199,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 3.989734411239624
    },
    {
      "auxiliary_loss_clip": 0.01174548,
      "auxiliary_loss_mlp": 0.01026171,
      "balance_loss_clip": 1.01114452,
      "balance_loss_mlp": 1.01851141,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 2.3674089327608723,
      "language_loss": 0.74328959,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.7652967,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 2.6990530490875244
    },
    {
      "auxiliary_loss_clip": 0.01173756,
      "auxiliary_loss_mlp": 0.01028924,
      "balance_loss_clip": 0.93389893,
      "balance_loss_mlp": 1.02174175,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 1.7266789031868994,
      "language_loss": 0.71823847,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.74026525,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.7026867866516113
    },
    {
      "auxiliary_loss_clip": 0.01171127,
      "auxiliary_loss_mlp": 0.01028899,
      "balance_loss_clip": 1.05007792,
      "balance_loss_mlp": 1.02191925,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 1.8679405997769958,
      "language_loss": 0.77732933,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.79932958,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 2.652538776397705
    },
    {
      "auxiliary_loss_clip": 0.01169381,
      "auxiliary_loss_mlp": 0.01029724,
      "balance_loss_clip": 0.97567153,
      "balance_loss_mlp": 1.02188349,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 1.9114424971154256,
      "language_loss": 0.76015389,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.7821449,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.68062162399292
    },
    {
      "auxiliary_loss_clip": 0.0116904,
      "auxiliary_loss_mlp": 0.01027202,
      "balance_loss_clip": 1.04694653,
      "balance_loss_mlp": 1.02030587,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 1.462996390645732,
      "language_loss": 0.74582869,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76779115,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 3.6866233348846436
    },
    {
      "auxiliary_loss_clip": 0.01177349,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 0.97213769,
      "balance_loss_mlp": 1.02219665,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 1.9186856201185936,
      "language_loss": 0.86904937,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89111722,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.7784807682037354
    },
    {
      "auxiliary_loss_clip": 0.01174873,
      "auxiliary_loss_mlp": 0.0102271,
      "balance_loss_clip": 0.97709143,
      "balance_loss_mlp": 1.01540291,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 1.6626566394784767,
      "language_loss": 0.73637533,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.75835121,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 3.5782103538513184
    },
    {
      "auxiliary_loss_clip": 0.0116983,
      "auxiliary_loss_mlp": 0.01029516,
      "balance_loss_clip": 0.9726488,
      "balance_loss_mlp": 1.02106404,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 1.844985516891751,
      "language_loss": 0.80596972,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.82796317,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 3.5492374897003174
    },
    {
      "auxiliary_loss_clip": 0.01165644,
      "auxiliary_loss_mlp": 0.01027453,
      "balance_loss_clip": 0.97153383,
      "balance_loss_mlp": 1.02061903,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 1.8968120528794141,
      "language_loss": 0.72160745,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74353838,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.6662251949310303
    },
    {
      "auxiliary_loss_clip": 0.01165943,
      "auxiliary_loss_mlp": 0.01031052,
      "balance_loss_clip": 0.97534883,
      "balance_loss_mlp": 1.0236845,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.4736920293246427,
      "language_loss": 0.70293814,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.72490811,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.806328535079956
    },
    {
      "auxiliary_loss_clip": 0.0116615,
      "auxiliary_loss_mlp": 0.01025905,
      "balance_loss_clip": 1.00982666,
      "balance_loss_mlp": 1.01807904,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 1.5335055092783683,
      "language_loss": 0.77468729,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.79660785,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.6100852489471436
    },
    {
      "auxiliary_loss_clip": 0.01163203,
      "auxiliary_loss_mlp": 0.01025959,
      "balance_loss_clip": 1.00840497,
      "balance_loss_mlp": 1.01815701,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 2.1950885948220726,
      "language_loss": 0.78702253,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.80891418,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.67533278465271
    },
    {
      "auxiliary_loss_clip": 0.01169413,
      "auxiliary_loss_mlp": 0.01026553,
      "balance_loss_clip": 1.0480684,
      "balance_loss_mlp": 1.01953721,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 2.3362947268263987,
      "language_loss": 0.7989288,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.8208884,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.5896360874176025
    },
    {
      "auxiliary_loss_clip": 0.01166647,
      "auxiliary_loss_mlp": 0.01024357,
      "balance_loss_clip": 1.01160586,
      "balance_loss_mlp": 1.01776755,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 1.5448244412277152,
      "language_loss": 0.8224982,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84440827,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.663482189178467
    },
    {
      "auxiliary_loss_clip": 0.01172211,
      "auxiliary_loss_mlp": 0.01025614,
      "balance_loss_clip": 0.97283125,
      "balance_loss_mlp": 1.0186522,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.5632360724866183,
      "language_loss": 0.83030862,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.85228682,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.6880712509155273
    },
    {
      "auxiliary_loss_clip": 0.01168538,
      "auxiliary_loss_mlp": 0.01030236,
      "balance_loss_clip": 0.97046417,
      "balance_loss_mlp": 1.02251768,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 1.7658966638857956,
      "language_loss": 0.63589895,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65788674,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.7260966300964355
    },
    {
      "auxiliary_loss_clip": 0.0117243,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.05174947,
      "balance_loss_mlp": 1.0188241,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 1.4801692185703939,
      "language_loss": 0.71510625,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73708755,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.86273193359375
    },
    {
      "auxiliary_loss_clip": 0.01175652,
      "auxiliary_loss_mlp": 0.011217,
      "balance_loss_clip": 0.97479606,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 1.8781945404794407,
      "language_loss": 0.82700741,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.84998095,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 2.667175531387329
    },
    {
      "auxiliary_loss_clip": 0.01177285,
      "auxiliary_loss_mlp": 0.01027938,
      "balance_loss_clip": 1.01166403,
      "balance_loss_mlp": 1.02020693,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 1.8436335924086036,
      "language_loss": 0.86804366,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.89009589,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 2.6355106830596924
    },
    {
      "auxiliary_loss_clip": 0.01170221,
      "auxiliary_loss_mlp": 0.01023138,
      "balance_loss_clip": 0.97214097,
      "balance_loss_mlp": 1.01597035,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 1.6520457527687513,
      "language_loss": 0.82714558,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.84907919,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 2.7476792335510254
    },
    {
      "auxiliary_loss_clip": 0.01173928,
      "auxiliary_loss_mlp": 0.01027749,
      "balance_loss_clip": 0.97354609,
      "balance_loss_mlp": 1.0202148,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 5.182148516454582,
      "language_loss": 0.79632312,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81833988,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.7183568477630615
    },
    {
      "auxiliary_loss_clip": 0.01168155,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 0.97190928,
      "balance_loss_mlp": 1.01993883,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 1.793171385620784,
      "language_loss": 0.70343161,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72539705,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.699187755584717
    },
    {
      "auxiliary_loss_clip": 0.01169932,
      "auxiliary_loss_mlp": 0.01030083,
      "balance_loss_clip": 0.96901429,
      "balance_loss_mlp": 1.02248597,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 2.805555169135198,
      "language_loss": 0.66169685,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.68369699,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 2.8143255710601807
    },
    {
      "auxiliary_loss_clip": 0.01167132,
      "auxiliary_loss_mlp": 0.0102246,
      "balance_loss_clip": 0.97507131,
      "balance_loss_mlp": 1.01571846,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 1.6877505271670195,
      "language_loss": 0.78246677,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80436271,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.669180393218994
    },
    {
      "auxiliary_loss_clip": 0.01158584,
      "auxiliary_loss_mlp": 0.01025823,
      "balance_loss_clip": 0.97349215,
      "balance_loss_mlp": 1.01880169,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 1.6070478391916454,
      "language_loss": 0.83084691,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85269094,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 2.6848108768463135
    },
    {
      "auxiliary_loss_clip": 0.01167957,
      "auxiliary_loss_mlp": 0.0103597,
      "balance_loss_clip": 1.00965309,
      "balance_loss_mlp": 1.02802515,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 2.025885007764381,
      "language_loss": 0.75925875,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.78129792,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 3.5570240020751953
    },
    {
      "auxiliary_loss_clip": 0.01166604,
      "auxiliary_loss_mlp": 0.01021133,
      "balance_loss_clip": 0.97318316,
      "balance_loss_mlp": 1.01428485,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.282328163671358,
      "language_loss": 0.76012057,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.78199792,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.65321683883667
    },
    {
      "auxiliary_loss_clip": 0.01072397,
      "auxiliary_loss_mlp": 0.01002621,
      "balance_loss_clip": 0.97736216,
      "balance_loss_mlp": 1.00111854,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7942883685654797,
      "language_loss": 0.57259929,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59334946,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.1945087909698486
    },
    {
      "auxiliary_loss_clip": 0.01173931,
      "auxiliary_loss_mlp": 0.01029093,
      "balance_loss_clip": 1.05004346,
      "balance_loss_mlp": 1.02186251,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 1.7421953053133394,
      "language_loss": 0.90589041,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92792058,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.6014606952667236
    },
    {
      "auxiliary_loss_clip": 0.01172877,
      "auxiliary_loss_mlp": 0.0103262,
      "balance_loss_clip": 0.97039533,
      "balance_loss_mlp": 1.02528298,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 1.8079653600584398,
      "language_loss": 0.77178907,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.79384398,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.6701765060424805
    },
    {
      "auxiliary_loss_clip": 0.01169668,
      "auxiliary_loss_mlp": 0.01021572,
      "balance_loss_clip": 0.97035265,
      "balance_loss_mlp": 1.01451468,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 1.9289558039272012,
      "language_loss": 0.76014829,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78206068,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 3.744518280029297
    },
    {
      "auxiliary_loss_clip": 0.01167235,
      "auxiliary_loss_mlp": 0.01024848,
      "balance_loss_clip": 1.00994062,
      "balance_loss_mlp": 1.01706409,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 1.6213961748707257,
      "language_loss": 0.79210126,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81402212,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.797499895095825
    },
    {
      "auxiliary_loss_clip": 0.01168093,
      "auxiliary_loss_mlp": 0.01024031,
      "balance_loss_clip": 1.01009583,
      "balance_loss_mlp": 1.01690519,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 2.5286855888358986,
      "language_loss": 0.69809139,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.72001266,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 3.634504556655884
    },
    {
      "auxiliary_loss_clip": 0.01171869,
      "auxiliary_loss_mlp": 0.010279,
      "balance_loss_clip": 0.97574604,
      "balance_loss_mlp": 1.02039623,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 1.848742223012662,
      "language_loss": 0.76431251,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.7863102,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 3.578697919845581
    },
    {
      "auxiliary_loss_clip": 0.01165906,
      "auxiliary_loss_mlp": 0.01024432,
      "balance_loss_clip": 1.00948906,
      "balance_loss_mlp": 1.01726413,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 1.7694094087346646,
      "language_loss": 0.86592126,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88782465,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.6851580142974854
    },
    {
      "auxiliary_loss_clip": 0.01178715,
      "auxiliary_loss_mlp": 0.01030918,
      "balance_loss_clip": 1.05329514,
      "balance_loss_mlp": 1.02354169,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 1.9584455846095052,
      "language_loss": 0.72087407,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74297029,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.537277936935425
    },
    {
      "auxiliary_loss_clip": 0.01167691,
      "auxiliary_loss_mlp": 0.01028087,
      "balance_loss_clip": 0.97519517,
      "balance_loss_mlp": 1.02101755,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 2.1744649141747012,
      "language_loss": 0.75562739,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77758521,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.665353775024414
    },
    {
      "auxiliary_loss_clip": 0.01175706,
      "auxiliary_loss_mlp": 0.01030654,
      "balance_loss_clip": 1.0522337,
      "balance_loss_mlp": 1.02253866,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 2.3389111750839295,
      "language_loss": 0.75626427,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77832788,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.70641827583313
    },
    {
      "auxiliary_loss_clip": 0.01170899,
      "auxiliary_loss_mlp": 0.01028725,
      "balance_loss_clip": 0.97213811,
      "balance_loss_mlp": 1.02173972,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 1.7824986862144814,
      "language_loss": 0.71444488,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73644114,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.7088747024536133
    },
    {
      "auxiliary_loss_clip": 0.01169794,
      "auxiliary_loss_mlp": 0.01025542,
      "balance_loss_clip": 0.97443074,
      "balance_loss_mlp": 1.01884866,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 1.7864431727821086,
      "language_loss": 0.8367036,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.85865688,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.6974055767059326
    },
    {
      "auxiliary_loss_clip": 0.01173773,
      "auxiliary_loss_mlp": 0.01027288,
      "balance_loss_clip": 1.05080581,
      "balance_loss_mlp": 1.02043343,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 3.3887756418030905,
      "language_loss": 0.73555005,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.75756067,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.6098763942718506
    },
    {
      "auxiliary_loss_clip": 0.01158703,
      "auxiliary_loss_mlp": 0.01025161,
      "balance_loss_clip": 0.97214311,
      "balance_loss_mlp": 1.01800871,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.8211509049046632,
      "language_loss": 0.68033648,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.70217514,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.715085506439209
    },
    {
      "auxiliary_loss_clip": 0.0106859,
      "auxiliary_loss_mlp": 0.00999955,
      "balance_loss_clip": 0.97667849,
      "balance_loss_mlp": 0.99832171,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.7883969468300963,
      "language_loss": 0.57469743,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59538293,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 3.2082626819610596
    },
    {
      "auxiliary_loss_clip": 0.01169745,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 0.97226977,
      "balance_loss_mlp": 1.02046692,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.244901412988615,
      "language_loss": 0.76155621,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.78353322,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 2.675795078277588
    },
    {
      "auxiliary_loss_clip": 0.01175908,
      "auxiliary_loss_mlp": 0.01032572,
      "balance_loss_clip": 0.89479893,
      "balance_loss_mlp": 1.02519274,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 1.6919897705342712,
      "language_loss": 0.71256042,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73464519,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.7585692405700684
    },
    {
      "auxiliary_loss_clip": 0.01170912,
      "auxiliary_loss_mlp": 0.01023763,
      "balance_loss_clip": 1.01187038,
      "balance_loss_mlp": 1.01638436,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.5213755577175525,
      "language_loss": 0.6515764,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67352313,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 2.7373359203338623
    },
    {
      "auxiliary_loss_clip": 0.01169605,
      "auxiliary_loss_mlp": 0.01023786,
      "balance_loss_clip": 1.01239097,
      "balance_loss_mlp": 1.01651978,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.6917616085299172,
      "language_loss": 0.76801229,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.7899462,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 2.719517707824707
    },
    {
      "auxiliary_loss_clip": 0.01167588,
      "auxiliary_loss_mlp": 0.01025948,
      "balance_loss_clip": 0.93313628,
      "balance_loss_mlp": 1.01799071,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 2.0556581281756072,
      "language_loss": 0.77907181,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.80100715,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 2.7013585567474365
    },
    {
      "auxiliary_loss_clip": 0.01163837,
      "auxiliary_loss_mlp": 0.01026195,
      "balance_loss_clip": 0.97058707,
      "balance_loss_mlp": 1.01851201,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 1.976057999585423,
      "language_loss": 0.79870266,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.82060301,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 2.7414562702178955
    },
    {
      "auxiliary_loss_clip": 0.01078597,
      "auxiliary_loss_mlp": 0.01000546,
      "balance_loss_clip": 0.9020462,
      "balance_loss_mlp": 0.99890101,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.8236426654244312,
      "language_loss": 0.5822078,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60299921,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 3.2954108715057373
    },
    {
      "auxiliary_loss_clip": 0.01170593,
      "auxiliary_loss_mlp": 0.01026685,
      "balance_loss_clip": 1.04973602,
      "balance_loss_mlp": 1.0192821,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 2.857492418773513,
      "language_loss": 0.67310822,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69508094,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 2.6239521503448486
    },
    {
      "auxiliary_loss_clip": 0.01175063,
      "auxiliary_loss_mlp": 0.01029626,
      "balance_loss_clip": 0.97491431,
      "balance_loss_mlp": 1.02218771,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 1.6563588280716481,
      "language_loss": 0.79512674,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.8171736,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 2.7596635818481445
    },
    {
      "auxiliary_loss_clip": 0.0115936,
      "auxiliary_loss_mlp": 0.0103078,
      "balance_loss_clip": 0.9709543,
      "balance_loss_mlp": 1.02295995,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 2.3436931825481655,
      "language_loss": 0.75175554,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.77365696,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 3.623995542526245
    },
    {
      "auxiliary_loss_clip": 0.01166162,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 0.97462004,
      "balance_loss_mlp": 1.02146208,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 1.8164313917848418,
      "language_loss": 0.82558608,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.84753686,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.7765262126922607
    },
    {
      "auxiliary_loss_clip": 0.01172532,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.05076146,
      "balance_loss_mlp": 1.02036881,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 1.7211631433848353,
      "language_loss": 0.78923976,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.81124562,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.601261615753174
    },
    {
      "auxiliary_loss_clip": 0.01163722,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 1.01060867,
      "balance_loss_mlp": 1.02118182,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.1645442605940572,
      "language_loss": 0.91705406,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.93897796,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.6084957122802734
    },
    {
      "auxiliary_loss_clip": 0.01162951,
      "auxiliary_loss_mlp": 0.01026671,
      "balance_loss_clip": 0.97102356,
      "balance_loss_mlp": 1.01930428,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 2.6315678025552063,
      "language_loss": 0.82311237,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84500855,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 3.6556310653686523
    },
    {
      "auxiliary_loss_clip": 0.01164187,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 0.97469926,
      "balance_loss_mlp": 1.0229919,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 2.0506503010728236,
      "language_loss": 0.70666647,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.72861409,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.6931121349334717
    },
    {
      "auxiliary_loss_clip": 0.01170572,
      "auxiliary_loss_mlp": 0.0102751,
      "balance_loss_clip": 1.00981927,
      "balance_loss_mlp": 1.01982057,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.253643345617657,
      "language_loss": 0.71635991,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.73834074,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 3.5368456840515137
    },
    {
      "auxiliary_loss_clip": 0.01163367,
      "auxiliary_loss_mlp": 0.01031047,
      "balance_loss_clip": 1.00784183,
      "balance_loss_mlp": 1.02291107,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 1.4858692515833114,
      "language_loss": 0.8181771,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.84012127,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 3.609910726547241
    },
    {
      "auxiliary_loss_clip": 0.01161306,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 0.93561941,
      "balance_loss_mlp": 1.01795626,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.5409156636897028,
      "language_loss": 0.84695387,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.86881912,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.758223056793213
    },
    {
      "auxiliary_loss_clip": 0.01165418,
      "auxiliary_loss_mlp": 0.01022872,
      "balance_loss_clip": 0.8989473,
      "balance_loss_mlp": 1.01562405,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 2.9258025480919643,
      "language_loss": 0.71621943,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.73810232,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.7176833152770996
    },
    {
      "auxiliary_loss_clip": 0.01163593,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 0.97361797,
      "balance_loss_mlp": 1.01965475,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 1.6318455071939066,
      "language_loss": 0.80890012,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.83080727,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.681588888168335
    },
    {
      "auxiliary_loss_clip": 0.01171973,
      "auxiliary_loss_mlp": 0.01028135,
      "balance_loss_clip": 1.00898516,
      "balance_loss_mlp": 1.02073181,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 2.124353579391665,
      "language_loss": 0.74981141,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77181256,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.588686227798462
    },
    {
      "auxiliary_loss_clip": 0.01165721,
      "auxiliary_loss_mlp": 0.01027193,
      "balance_loss_clip": 0.93208051,
      "balance_loss_mlp": 1.0199213,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 1.7522997005132805,
      "language_loss": 0.80575562,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.82768476,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.7357351779937744
    },
    {
      "auxiliary_loss_clip": 0.01162601,
      "auxiliary_loss_mlp": 0.01028918,
      "balance_loss_clip": 0.97309858,
      "balance_loss_mlp": 1.02165771,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 1.819666846796521,
      "language_loss": 0.88787699,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.90979218,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.732231855392456
    },
    {
      "auxiliary_loss_clip": 0.0115694,
      "auxiliary_loss_mlp": 0.01022752,
      "balance_loss_clip": 0.93357742,
      "balance_loss_mlp": 1.01529527,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 1.6347410546052752,
      "language_loss": 0.82913506,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.85093188,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.650526523590088
    },
    {
      "auxiliary_loss_clip": 0.01173591,
      "auxiliary_loss_mlp": 0.01026813,
      "balance_loss_clip": 1.05032277,
      "balance_loss_mlp": 1.01950574,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 1.6385025588939435,
      "language_loss": 0.80213243,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.8241365,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.6835505962371826
    },
    {
      "auxiliary_loss_clip": 0.01167388,
      "auxiliary_loss_mlp": 0.01030915,
      "balance_loss_clip": 0.93376517,
      "balance_loss_mlp": 1.02396846,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 4.021898268686976,
      "language_loss": 0.78171575,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.80369884,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.691572427749634
    },
    {
      "auxiliary_loss_clip": 0.01172253,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.01124322,
      "balance_loss_mlp": 1.01876593,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 1.6674976761077636,
      "language_loss": 0.77229649,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79428786,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 2.6854259967803955
    },
    {
      "auxiliary_loss_clip": 0.01162796,
      "auxiliary_loss_mlp": 0.01123086,
      "balance_loss_clip": 0.9326911,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 2.097120597200302,
      "language_loss": 0.77987558,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.80273438,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 2.676046371459961
    },
    {
      "auxiliary_loss_clip": 0.01160571,
      "auxiliary_loss_mlp": 0.01024238,
      "balance_loss_clip": 0.97080451,
      "balance_loss_mlp": 1.01693916,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.3237227982301976,
      "language_loss": 0.80850339,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83035147,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 2.743741750717163
    },
    {
      "auxiliary_loss_clip": 0.01068293,
      "auxiliary_loss_mlp": 0.01002312,
      "balance_loss_clip": 1.01388752,
      "balance_loss_mlp": 1.00075078,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.781158858008484,
      "language_loss": 0.60631371,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62701976,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.2781553268432617
    },
    {
      "auxiliary_loss_clip": 0.01168836,
      "auxiliary_loss_mlp": 0.01026875,
      "balance_loss_clip": 0.97706348,
      "balance_loss_mlp": 1.01995778,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 2.615278354060527,
      "language_loss": 0.72866988,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.75062692,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 2.6816482543945312
    },
    {
      "auxiliary_loss_clip": 0.01174968,
      "auxiliary_loss_mlp": 0.01122932,
      "balance_loss_clip": 0.85645932,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.2471045660608,
      "language_loss": 0.71227193,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.73525095,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 2.862395763397217
    },
    {
      "auxiliary_loss_clip": 0.01173895,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.01021266,
      "balance_loss_mlp": 1.01841736,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 1.7506186919713371,
      "language_loss": 0.79979837,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82179183,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 2.9608728885650635
    },
    {
      "auxiliary_loss_clip": 0.01173315,
      "auxiliary_loss_mlp": 0.01028076,
      "balance_loss_clip": 0.89721334,
      "balance_loss_mlp": 1.02009463,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 1.838831790264973,
      "language_loss": 0.72533357,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.74734747,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 2.776017904281616
    },
    {
      "auxiliary_loss_clip": 0.01168901,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 0.94048285,
      "balance_loss_mlp": 1.01796651,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 1.881121728124457,
      "language_loss": 0.66134882,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68329465,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 3.697387933731079
    },
    {
      "auxiliary_loss_clip": 0.01173003,
      "auxiliary_loss_mlp": 0.01029975,
      "balance_loss_clip": 0.97464591,
      "balance_loss_mlp": 1.02247643,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 3.289901846442159,
      "language_loss": 0.70051515,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.72254491,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 2.671170711517334
    },
    {
      "auxiliary_loss_clip": 0.01159553,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 0.89493501,
      "balance_loss_mlp": 1.02368307,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.6274641485251453,
      "language_loss": 0.74412328,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.76603663,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.948216676712036
    },
    {
      "auxiliary_loss_clip": 0.01168566,
      "auxiliary_loss_mlp": 0.01024654,
      "balance_loss_clip": 1.01090288,
      "balance_loss_mlp": 1.01700115,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 1.9778871140150691,
      "language_loss": 0.79641962,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81835186,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.6274309158325195
    },
    {
      "auxiliary_loss_clip": 0.01177111,
      "auxiliary_loss_mlp": 0.01024548,
      "balance_loss_clip": 1.01092696,
      "balance_loss_mlp": 1.01725876,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 1.5910605650244694,
      "language_loss": 0.73367655,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.7556932,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.6772048473358154
    },
    {
      "auxiliary_loss_clip": 0.01150929,
      "auxiliary_loss_mlp": 0.01023383,
      "balance_loss_clip": 0.81934547,
      "balance_loss_mlp": 1.01630807,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 6.349920378318319,
      "language_loss": 0.85614753,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.87789065,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 3.741163969039917
    },
    {
      "auxiliary_loss_clip": 0.01169828,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.01045251,
      "balance_loss_mlp": 1.01789176,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 1.9900499957939326,
      "language_loss": 0.88324034,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.9051891,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.7104341983795166
    },
    {
      "auxiliary_loss_clip": 0.01173644,
      "auxiliary_loss_mlp": 0.01122746,
      "balance_loss_clip": 0.9729172,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.7616398376649378,
      "language_loss": 0.74229538,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.76525927,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 3.622174024581909
    },
    {
      "auxiliary_loss_clip": 0.0116299,
      "auxiliary_loss_mlp": 0.01024432,
      "balance_loss_clip": 0.97308463,
      "balance_loss_mlp": 1.01676679,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 1.7766426550925385,
      "language_loss": 0.66218597,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68406016,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.6710779666900635
    },
    {
      "auxiliary_loss_clip": 0.01175843,
      "auxiliary_loss_mlp": 0.01026751,
      "balance_loss_clip": 1.01365089,
      "balance_loss_mlp": 1.0192349,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 2.60946760296633,
      "language_loss": 0.82222772,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.84425366,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.6860201358795166
    },
    {
      "auxiliary_loss_clip": 0.01174142,
      "auxiliary_loss_mlp": 0.01025181,
      "balance_loss_clip": 0.97489285,
      "balance_loss_mlp": 1.01804018,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 2.318533750832802,
      "language_loss": 0.79307759,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81507087,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.619142770767212
    },
    {
      "auxiliary_loss_clip": 0.01172425,
      "auxiliary_loss_mlp": 0.01027732,
      "balance_loss_clip": 0.97128278,
      "balance_loss_mlp": 1.02086258,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 2.177568775158125,
      "language_loss": 0.84791517,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.86991668,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.7820191383361816
    },
    {
      "auxiliary_loss_clip": 0.01172796,
      "auxiliary_loss_mlp": 0.01122499,
      "balance_loss_clip": 1.01234424,
      "balance_loss_mlp": 0.0,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.6307965817850631,
      "language_loss": 0.87007535,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.89302826,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.7277889251708984
    },
    {
      "auxiliary_loss_clip": 0.01174297,
      "auxiliary_loss_mlp": 0.0102804,
      "balance_loss_clip": 1.05057967,
      "balance_loss_mlp": 1.01992774,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 1.8107880713866467,
      "language_loss": 0.69154847,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.71357179,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.685330390930176
    },
    {
      "auxiliary_loss_clip": 0.01171919,
      "auxiliary_loss_mlp": 0.01026316,
      "balance_loss_clip": 1.01493406,
      "balance_loss_mlp": 1.01911628,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 1.855653394508742,
      "language_loss": 0.61854666,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.64052904,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.8005893230438232
    },
    {
      "auxiliary_loss_clip": 0.01170184,
      "auxiliary_loss_mlp": 0.01024517,
      "balance_loss_clip": 1.01102018,
      "balance_loss_mlp": 1.01749527,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 2.1328725418052135,
      "language_loss": 0.70044863,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.72239566,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.677809000015259
    },
    {
      "auxiliary_loss_clip": 0.01154754,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 0.8929792,
      "balance_loss_mlp": 1.01947236,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.4469548120213547,
      "language_loss": 0.83093703,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.85275018,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.7677581310272217
    },
    {
      "auxiliary_loss_clip": 0.01161308,
      "auxiliary_loss_mlp": 0.01123034,
      "balance_loss_clip": 0.85660285,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 1.7103413446590028,
      "language_loss": 0.76817405,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.79101741,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.7403719425201416
    },
    {
      "auxiliary_loss_clip": 0.01178051,
      "auxiliary_loss_mlp": 0.01029411,
      "balance_loss_clip": 0.97612375,
      "balance_loss_mlp": 1.02160871,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 2.5639164810286617,
      "language_loss": 0.72665828,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.74873292,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 2.6679186820983887
    },
    {
      "auxiliary_loss_clip": 0.01169112,
      "auxiliary_loss_mlp": 0.01027957,
      "balance_loss_clip": 1.00960374,
      "balance_loss_mlp": 1.02038074,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 2.541170488111723,
      "language_loss": 0.88814008,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.91011077,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 2.58976149559021
    },
    {
      "auxiliary_loss_clip": 0.01155883,
      "auxiliary_loss_mlp": 0.01026519,
      "balance_loss_clip": 0.85636258,
      "balance_loss_mlp": 1.01950383,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.4477185632211904,
      "language_loss": 0.86174226,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88356626,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 2.9317967891693115
    },
    {
      "auxiliary_loss_clip": 0.01170926,
      "auxiliary_loss_mlp": 0.01020983,
      "balance_loss_clip": 1.01271558,
      "balance_loss_mlp": 1.01418257,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 2.2090537608701593,
      "language_loss": 0.77303147,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.7949506,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.6159582138061523
    },
    {
      "auxiliary_loss_clip": 0.01170218,
      "auxiliary_loss_mlp": 0.01029511,
      "balance_loss_clip": 1.05010688,
      "balance_loss_mlp": 1.02221251,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 1.605494202489554,
      "language_loss": 0.71628386,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.73828119,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 2.6602354049682617
    },
    {
      "auxiliary_loss_clip": 0.01177075,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.01222825,
      "balance_loss_mlp": 1.01750636,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 1.6032144572970708,
      "language_loss": 0.73505092,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75706756,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.6422648429870605
    },
    {
      "auxiliary_loss_clip": 0.01084014,
      "auxiliary_loss_mlp": 0.01001103,
      "balance_loss_clip": 0.95353186,
      "balance_loss_mlp": 0.99938625,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.836756107084329,
      "language_loss": 0.60917735,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.63002849,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 3.287794589996338
    },
    {
      "auxiliary_loss_clip": 0.01089805,
      "auxiliary_loss_mlp": 0.01000309,
      "balance_loss_clip": 0.88354844,
      "balance_loss_mlp": 0.99874717,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8511215956885204,
      "language_loss": 0.57886875,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59976989,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 3.210911273956299
    },
    {
      "auxiliary_loss_clip": 0.01158544,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 0.9332298,
      "balance_loss_mlp": 1.02177179,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.0498866553764827,
      "language_loss": 0.75300533,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77488387,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 3.74656081199646
    },
    {
      "auxiliary_loss_clip": 0.01178098,
      "auxiliary_loss_mlp": 0.01122566,
      "balance_loss_clip": 0.97688907,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 2.6503567501687098,
      "language_loss": 0.7083019,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.73130846,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.6938540935516357
    },
    {
      "auxiliary_loss_clip": 0.01173535,
      "auxiliary_loss_mlp": 0.01030201,
      "balance_loss_clip": 0.97493649,
      "balance_loss_mlp": 1.02267325,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 1.7668153993644513,
      "language_loss": 0.74189246,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76392978,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 2.8336353302001953
    },
    {
      "auxiliary_loss_clip": 0.01177753,
      "auxiliary_loss_mlp": 0.01024465,
      "balance_loss_clip": 0.93530083,
      "balance_loss_mlp": 1.01752067,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 2.108132268764228,
      "language_loss": 0.78680897,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80883116,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.7559008598327637
    },
    {
      "auxiliary_loss_clip": 0.01166563,
      "auxiliary_loss_mlp": 0.01029194,
      "balance_loss_clip": 0.97237551,
      "balance_loss_mlp": 1.02169538,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.743270818347703,
      "language_loss": 0.6946072,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.71656477,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.8307430744171143
    },
    {
      "auxiliary_loss_clip": 0.01069135,
      "auxiliary_loss_mlp": 0.01005332,
      "balance_loss_clip": 1.01444125,
      "balance_loss_mlp": 1.00374627,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 1.5174722751227943,
      "language_loss": 0.51189142,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53263611,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.3678927421569824
    },
    {
      "auxiliary_loss_clip": 0.01158168,
      "auxiliary_loss_mlp": 0.01030016,
      "balance_loss_clip": 0.93150014,
      "balance_loss_mlp": 1.02264595,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.6948225908697547,
      "language_loss": 0.70692092,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72880274,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 3.6640286445617676
    },
    {
      "auxiliary_loss_clip": 0.01171091,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 0.90133113,
      "balance_loss_mlp": 1.02145028,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 1.981436932787451,
      "language_loss": 0.71016192,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.73215711,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 4.521790504455566
    },
    {
      "auxiliary_loss_clip": 0.01176112,
      "auxiliary_loss_mlp": 0.01027227,
      "balance_loss_clip": 0.97392523,
      "balance_loss_mlp": 1.02021503,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 1.5756777844494658,
      "language_loss": 0.77588665,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.79792011,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.6688072681427
    },
    {
      "auxiliary_loss_clip": 0.01170944,
      "auxiliary_loss_mlp": 0.01029101,
      "balance_loss_clip": 1.01307118,
      "balance_loss_mlp": 1.021981,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 2.3996157865982215,
      "language_loss": 0.84965694,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.87165731,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.637018918991089
    },
    {
      "auxiliary_loss_clip": 0.01174908,
      "auxiliary_loss_mlp": 0.01027727,
      "balance_loss_clip": 1.0497303,
      "balance_loss_mlp": 1.01979327,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.6593610644060053,
      "language_loss": 0.74490333,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76692975,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.598961591720581
    },
    {
      "auxiliary_loss_clip": 0.01161665,
      "auxiliary_loss_mlp": 0.01024086,
      "balance_loss_clip": 1.00921583,
      "balance_loss_mlp": 1.01727295,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 1.6498413938421788,
      "language_loss": 0.78661752,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.80847502,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.6911680698394775
    },
    {
      "auxiliary_loss_clip": 0.01167641,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 0.9309603,
      "balance_loss_mlp": 1.02054906,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.5919160508964596,
      "language_loss": 0.84300351,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86496073,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.8394083976745605
    },
    {
      "auxiliary_loss_clip": 0.01175318,
      "auxiliary_loss_mlp": 0.01029665,
      "balance_loss_clip": 1.01743877,
      "balance_loss_mlp": 1.02230418,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 1.96265967094198,
      "language_loss": 0.6710304,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69308031,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.6831839084625244
    },
    {
      "auxiliary_loss_clip": 0.01171203,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.05071425,
      "balance_loss_mlp": 1.0190022,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.5667323494368979,
      "language_loss": 0.58823007,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61020398,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.806478977203369
    },
    {
      "auxiliary_loss_clip": 0.01168842,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 0.97377992,
      "balance_loss_mlp": 1.01883376,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 1.8310216267563213,
      "language_loss": 0.91136098,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93331647,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.813859701156616
    },
    {
      "auxiliary_loss_clip": 0.01177205,
      "auxiliary_loss_mlp": 0.01123216,
      "balance_loss_clip": 1.01661193,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 1.6156622882325755,
      "language_loss": 0.76993591,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.79294008,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.723015308380127
    },
    {
      "auxiliary_loss_clip": 0.01144627,
      "auxiliary_loss_mlp": 0.01027932,
      "balance_loss_clip": 0.77230799,
      "balance_loss_mlp": 1.02049351,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 2.084759566670751,
      "language_loss": 0.76635647,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.788082,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 2.9589200019836426
    },
    {
      "auxiliary_loss_clip": 0.01176772,
      "auxiliary_loss_mlp": 0.01025527,
      "balance_loss_clip": 1.01309729,
      "balance_loss_mlp": 1.01799893,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 1.8957683201636908,
      "language_loss": 0.77965367,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80167663,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 3.240668296813965
    },
    {
      "auxiliary_loss_clip": 0.01149545,
      "auxiliary_loss_mlp": 0.01023553,
      "balance_loss_clip": 0.9258914,
      "balance_loss_mlp": 1.01660895,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 1.8300587253700602,
      "language_loss": 0.72808355,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.74981451,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 2.8121705055236816
    },
    {
      "auxiliary_loss_clip": 0.01162524,
      "auxiliary_loss_mlp": 0.01121836,
      "balance_loss_clip": 0.86066788,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 1.9401159847603493,
      "language_loss": 0.75185388,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.77469754,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 2.8073208332061768
    },
    {
      "auxiliary_loss_clip": 0.01163388,
      "auxiliary_loss_mlp": 0.01030256,
      "balance_loss_clip": 0.93371463,
      "balance_loss_mlp": 1.02319312,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 1.731475882080453,
      "language_loss": 0.67074561,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.69268203,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 2.7982733249664307
    },
    {
      "auxiliary_loss_clip": 0.0116463,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.00985384,
      "balance_loss_mlp": 1.01926577,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 1.6003825836281782,
      "language_loss": 0.75222766,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77414042,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 2.6258270740509033
    },
    {
      "auxiliary_loss_clip": 0.01175278,
      "auxiliary_loss_mlp": 0.01023162,
      "balance_loss_clip": 1.05332303,
      "balance_loss_mlp": 1.01626527,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.5668857827691793,
      "language_loss": 0.749466,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.7714504,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.6662561893463135
    },
    {
      "auxiliary_loss_clip": 0.01156022,
      "auxiliary_loss_mlp": 0.01123269,
      "balance_loss_clip": 0.89539123,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 2.9366996072770384,
      "language_loss": 0.72450036,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.74729329,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 2.7111101150512695
    },
    {
      "auxiliary_loss_clip": 0.01158788,
      "auxiliary_loss_mlp": 0.0102981,
      "balance_loss_clip": 0.93386841,
      "balance_loss_mlp": 1.02139974,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 1.791093709821655,
      "language_loss": 0.71062887,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73251486,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 2.7876901626586914
    },
    {
      "auxiliary_loss_clip": 0.01167342,
      "auxiliary_loss_mlp": 0.0102938,
      "balance_loss_clip": 0.93115652,
      "balance_loss_mlp": 1.02157199,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 2.172231731000855,
      "language_loss": 0.76983011,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.79179734,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 4.057806015014648
    },
    {
      "auxiliary_loss_clip": 0.01173517,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 0.97194636,
      "balance_loss_mlp": 1.01927209,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 1.6441276364580137,
      "language_loss": 0.7403996,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.76240063,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.776503086090088
    },
    {
      "auxiliary_loss_clip": 0.01169462,
      "auxiliary_loss_mlp": 0.01024371,
      "balance_loss_clip": 0.97578788,
      "balance_loss_mlp": 1.01741469,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 1.8650580755921193,
      "language_loss": 0.81352365,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.83546197,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 2.8114101886749268
    },
    {
      "auxiliary_loss_clip": 0.01165265,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 0.96969646,
      "balance_loss_mlp": 1.01985502,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 2.3683299052668225,
      "language_loss": 0.73466468,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.75658935,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.800865411758423
    },
    {
      "auxiliary_loss_clip": 0.01174396,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.05153465,
      "balance_loss_mlp": 1.01910949,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 1.770076622670259,
      "language_loss": 0.69903183,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.7210452,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.7441906929016113
    },
    {
      "auxiliary_loss_clip": 0.01168002,
      "auxiliary_loss_mlp": 0.01024639,
      "balance_loss_clip": 1.01218486,
      "balance_loss_mlp": 1.01748681,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 2.164341818551392,
      "language_loss": 0.75417578,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77610219,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 3.6443674564361572
    },
    {
      "auxiliary_loss_clip": 0.01156987,
      "auxiliary_loss_mlp": 0.01021047,
      "balance_loss_clip": 0.97205049,
      "balance_loss_mlp": 1.01429045,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 1.6973813241441724,
      "language_loss": 0.70465803,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.7264384,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.6465513706207275
    },
    {
      "auxiliary_loss_clip": 0.01171374,
      "auxiliary_loss_mlp": 0.0102542,
      "balance_loss_clip": 0.93677711,
      "balance_loss_mlp": 1.01770711,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 1.8398232566643073,
      "language_loss": 0.62234229,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.64431024,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 3.8020007610321045
    },
    {
      "auxiliary_loss_clip": 0.01176229,
      "auxiliary_loss_mlp": 0.01033777,
      "balance_loss_clip": 0.93457907,
      "balance_loss_mlp": 1.0258373,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 1.8864215402609985,
      "language_loss": 0.77485114,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79695117,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 3.675651788711548
    },
    {
      "auxiliary_loss_clip": 0.01172838,
      "auxiliary_loss_mlp": 0.01024353,
      "balance_loss_clip": 0.97249645,
      "balance_loss_mlp": 1.01758218,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.9950093654419827,
      "language_loss": 0.81414926,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.8361212,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.6907148361206055
    },
    {
      "auxiliary_loss_clip": 0.01174196,
      "auxiliary_loss_mlp": 0.010297,
      "balance_loss_clip": 1.05148911,
      "balance_loss_mlp": 1.02233863,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 2.750267467236352,
      "language_loss": 0.80284733,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82488632,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.591557502746582
    },
    {
      "auxiliary_loss_clip": 0.01169733,
      "auxiliary_loss_mlp": 0.01026717,
      "balance_loss_clip": 1.01190436,
      "balance_loss_mlp": 1.01927781,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 1.884058777130553,
      "language_loss": 0.72916996,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.7511344,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.7457847595214844
    },
    {
      "auxiliary_loss_clip": 0.01171293,
      "auxiliary_loss_mlp": 0.01033058,
      "balance_loss_clip": 1.01190555,
      "balance_loss_mlp": 1.02568173,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.057516992787506,
      "language_loss": 0.75359476,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77563828,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.703245162963867
    },
    {
      "auxiliary_loss_clip": 0.01171764,
      "auxiliary_loss_mlp": 0.01024055,
      "balance_loss_clip": 1.01038742,
      "balance_loss_mlp": 1.01732898,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 2.52238292789011,
      "language_loss": 0.8405447,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.86250293,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.6350066661834717
    },
    {
      "auxiliary_loss_clip": 0.01168907,
      "auxiliary_loss_mlp": 0.01027895,
      "balance_loss_clip": 0.97524995,
      "balance_loss_mlp": 1.02060282,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 2.0921720017197147,
      "language_loss": 0.71327657,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73524463,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.7389261722564697
    },
    {
      "auxiliary_loss_clip": 0.01175443,
      "auxiliary_loss_mlp": 0.01030239,
      "balance_loss_clip": 0.97452754,
      "balance_loss_mlp": 1.02231133,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.5185628099374053,
      "language_loss": 0.73520911,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.75726593,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.7142696380615234
    },
    {
      "auxiliary_loss_clip": 0.01176684,
      "auxiliary_loss_mlp": 0.01022338,
      "balance_loss_clip": 1.0121491,
      "balance_loss_mlp": 1.01480412,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 1.7640312458526837,
      "language_loss": 0.74651122,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76850152,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.703012466430664
    },
    {
      "auxiliary_loss_clip": 0.01170764,
      "auxiliary_loss_mlp": 0.0102276,
      "balance_loss_clip": 1.04821444,
      "balance_loss_mlp": 1.0151422,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 2.2435883148697395,
      "language_loss": 0.85249031,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87442553,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.67287278175354
    },
    {
      "auxiliary_loss_clip": 0.01163793,
      "auxiliary_loss_mlp": 0.01032394,
      "balance_loss_clip": 0.93373716,
      "balance_loss_mlp": 1.0247705,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.1757492914789207,
      "language_loss": 0.85373265,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.87569451,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.7620763778686523
    },
    {
      "auxiliary_loss_clip": 0.01182654,
      "auxiliary_loss_mlp": 0.01028145,
      "balance_loss_clip": 0.97626412,
      "balance_loss_mlp": 1.0207752,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 1.9121272635033535,
      "language_loss": 0.78597248,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.80808043,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 2.7486302852630615
    },
    {
      "auxiliary_loss_clip": 0.01175835,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.0122999,
      "balance_loss_mlp": 1.01815653,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 1.6335450145571684,
      "language_loss": 0.72081411,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.74282819,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 2.720386028289795
    },
    {
      "auxiliary_loss_clip": 0.01177311,
      "auxiliary_loss_mlp": 0.01029069,
      "balance_loss_clip": 0.9773227,
      "balance_loss_mlp": 1.02119827,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 1.801287807331537,
      "language_loss": 0.75996321,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78202701,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.6923959255218506
    },
    {
      "auxiliary_loss_clip": 0.01152383,
      "auxiliary_loss_mlp": 0.0112254,
      "balance_loss_clip": 0.89204252,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 1.835102466432563,
      "language_loss": 0.84318155,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.8659308,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 2.6915438175201416
    },
    {
      "auxiliary_loss_clip": 0.01172912,
      "auxiliary_loss_mlp": 0.0102491,
      "balance_loss_clip": 0.97517133,
      "balance_loss_mlp": 1.01731622,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.406022613210267,
      "language_loss": 0.76422167,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78619993,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.6984620094299316
    },
    {
      "auxiliary_loss_clip": 0.0117324,
      "auxiliary_loss_mlp": 0.01028618,
      "balance_loss_clip": 1.04872978,
      "balance_loss_mlp": 1.021191,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 1.6119197147476583,
      "language_loss": 0.77547181,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.79749042,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 2.654129981994629
    },
    {
      "auxiliary_loss_clip": 0.01170678,
      "auxiliary_loss_mlp": 0.01030627,
      "balance_loss_clip": 0.86066914,
      "balance_loss_mlp": 1.02289009,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 2.4430061838072996,
      "language_loss": 0.90343988,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.92545289,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 2.7159860134124756
    },
    {
      "auxiliary_loss_clip": 0.01177846,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 0.97310412,
      "balance_loss_mlp": 1.02041268,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.1097464253096225,
      "language_loss": 0.8206141,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84267211,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 3.681229591369629
    },
    {
      "auxiliary_loss_clip": 0.01171975,
      "auxiliary_loss_mlp": 0.01028687,
      "balance_loss_clip": 0.97073013,
      "balance_loss_mlp": 1.02124834,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 2.014748275551256,
      "language_loss": 0.7144717,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.73647827,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 2.684199810028076
    },
    {
      "auxiliary_loss_clip": 0.01168953,
      "auxiliary_loss_mlp": 0.01029578,
      "balance_loss_clip": 0.89337528,
      "balance_loss_mlp": 1.02177584,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 1.6845030729054675,
      "language_loss": 0.70580631,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.72779161,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.8000335693359375
    },
    {
      "auxiliary_loss_clip": 0.01171098,
      "auxiliary_loss_mlp": 0.01031013,
      "balance_loss_clip": 0.9719224,
      "balance_loss_mlp": 1.02398276,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 1.9013233096861357,
      "language_loss": 0.78327835,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80529952,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.6556384563446045
    },
    {
      "auxiliary_loss_clip": 0.01171348,
      "auxiliary_loss_mlp": 0.01024461,
      "balance_loss_clip": 0.97747934,
      "balance_loss_mlp": 1.01692986,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 2.2153442298496113,
      "language_loss": 0.66038167,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68233967,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.6778433322906494
    },
    {
      "auxiliary_loss_clip": 0.01180964,
      "auxiliary_loss_mlp": 0.01029696,
      "balance_loss_clip": 0.93691003,
      "balance_loss_mlp": 1.02153039,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 1.8254483075805887,
      "language_loss": 0.74098051,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.76308709,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 3.6838760375976562
    },
    {
      "auxiliary_loss_clip": 0.01172945,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.01521814,
      "balance_loss_mlp": 1.02035379,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 2.031132943761988,
      "language_loss": 0.67664945,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.69865167,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.6641032695770264
    },
    {
      "auxiliary_loss_clip": 0.01166824,
      "auxiliary_loss_mlp": 0.0102472,
      "balance_loss_clip": 1.01127291,
      "balance_loss_mlp": 1.01753759,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.637958211180795,
      "language_loss": 0.72239506,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.74431044,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 3.5495119094848633
    },
    {
      "auxiliary_loss_clip": 0.0116359,
      "auxiliary_loss_mlp": 0.01029604,
      "balance_loss_clip": 0.97245085,
      "balance_loss_mlp": 1.02194452,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 1.8092131227228347,
      "language_loss": 0.80479336,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.82672524,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 3.634117603302002
    },
    {
      "auxiliary_loss_clip": 0.01179432,
      "auxiliary_loss_mlp": 0.01033743,
      "balance_loss_clip": 1.01318216,
      "balance_loss_mlp": 1.02570188,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 2.6406079194185437,
      "language_loss": 0.77620387,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79833567,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.6164722442626953
    },
    {
      "auxiliary_loss_clip": 0.01164836,
      "auxiliary_loss_mlp": 0.01027272,
      "balance_loss_clip": 0.97286958,
      "balance_loss_mlp": 1.01979792,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 1.8396213053144865,
      "language_loss": 0.69614643,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71806753,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.711641550064087
    },
    {
      "auxiliary_loss_clip": 0.01182094,
      "auxiliary_loss_mlp": 0.01033186,
      "balance_loss_clip": 0.93658811,
      "balance_loss_mlp": 1.02505302,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 1.9305192476339044,
      "language_loss": 0.73149371,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75364649,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.8176238536834717
    },
    {
      "auxiliary_loss_clip": 0.01176538,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 1.05098426,
      "balance_loss_mlp": 1.01796174,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 1.6951273122277561,
      "language_loss": 0.75531423,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.77734077,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.6652424335479736
    },
    {
      "auxiliary_loss_clip": 0.01169225,
      "auxiliary_loss_mlp": 0.01030865,
      "balance_loss_clip": 1.0119133,
      "balance_loss_mlp": 1.02370644,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 5.307970214042332,
      "language_loss": 0.80171007,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82371104,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.6720194816589355
    },
    {
      "auxiliary_loss_clip": 0.01174342,
      "auxiliary_loss_mlp": 0.01021851,
      "balance_loss_clip": 0.97485799,
      "balance_loss_mlp": 1.01453769,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 1.484075597642645,
      "language_loss": 0.70234942,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72431129,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.712054967880249
    },
    {
      "auxiliary_loss_clip": 0.01161356,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 0.85359567,
      "balance_loss_mlp": 1.02189696,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 1.5813507043044366,
      "language_loss": 0.78009319,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80199981,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.8245487213134766
    },
    {
      "auxiliary_loss_clip": 0.01172885,
      "auxiliary_loss_mlp": 0.01028279,
      "balance_loss_clip": 0.9734208,
      "balance_loss_mlp": 1.02092957,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 2.4911493980279062,
      "language_loss": 0.85274875,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.87476039,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.669036626815796
    },
    {
      "auxiliary_loss_clip": 0.01174921,
      "auxiliary_loss_mlp": 0.01027415,
      "balance_loss_clip": 1.01433325,
      "balance_loss_mlp": 1.01967216,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 1.6908961135025182,
      "language_loss": 0.74613202,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.76815546,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.586435079574585
    },
    {
      "auxiliary_loss_clip": 0.01174472,
      "auxiliary_loss_mlp": 0.0102987,
      "balance_loss_clip": 1.01289701,
      "balance_loss_mlp": 1.02252948,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 2.1143751653177087,
      "language_loss": 0.83947134,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86151475,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.663515567779541
    },
    {
      "auxiliary_loss_clip": 0.01164289,
      "auxiliary_loss_mlp": 0.01029701,
      "balance_loss_clip": 0.93730485,
      "balance_loss_mlp": 1.02205324,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 2.0013543806319958,
      "language_loss": 0.68094963,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.70288956,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 2.8330795764923096
    },
    {
      "auxiliary_loss_clip": 0.01177083,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 0.97195977,
      "balance_loss_mlp": 1.02070594,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 2.177915637932996,
      "language_loss": 0.71260989,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.7346648,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.7487235069274902
    },
    {
      "auxiliary_loss_clip": 0.01172791,
      "auxiliary_loss_mlp": 0.0102316,
      "balance_loss_clip": 1.05123818,
      "balance_loss_mlp": 1.01594758,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 1.887128796729829,
      "language_loss": 0.60692507,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.62888455,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 2.7548558712005615
    },
    {
      "auxiliary_loss_clip": 0.01165515,
      "auxiliary_loss_mlp": 0.0102638,
      "balance_loss_clip": 0.97138119,
      "balance_loss_mlp": 1.01874471,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 1.8175057145683415,
      "language_loss": 0.82686806,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.84878695,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 2.7195546627044678
    },
    {
      "auxiliary_loss_clip": 0.01177519,
      "auxiliary_loss_mlp": 0.01028147,
      "balance_loss_clip": 1.05267358,
      "balance_loss_mlp": 1.02044046,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.7495970124576123,
      "language_loss": 0.79481971,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.81687629,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.6946768760681152
    },
    {
      "auxiliary_loss_clip": 0.01169346,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 0.97208816,
      "balance_loss_mlp": 1.01978004,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.0125825390545606,
      "language_loss": 0.75428653,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.77625197,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 2.6322410106658936
    },
    {
      "auxiliary_loss_clip": 0.01178225,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 0.97345531,
      "balance_loss_mlp": 1.01901364,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.0567979758603827,
      "language_loss": 0.71550107,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.73755062,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 2.770723342895508
    },
    {
      "auxiliary_loss_clip": 0.01173246,
      "auxiliary_loss_mlp": 0.01023862,
      "balance_loss_clip": 1.04964936,
      "balance_loss_mlp": 1.01675081,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.0702118944462775,
      "language_loss": 0.84432709,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.8662982,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 3.4525790214538574
    },
    {
      "auxiliary_loss_clip": 0.01170805,
      "auxiliary_loss_mlp": 0.01030244,
      "balance_loss_clip": 1.0129571,
      "balance_loss_mlp": 1.02275503,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 1.7924639844901717,
      "language_loss": 0.8767457,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.89875621,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 2.592158555984497
    },
    {
      "auxiliary_loss_clip": 0.01170243,
      "auxiliary_loss_mlp": 0.01023654,
      "balance_loss_clip": 0.97321224,
      "balance_loss_mlp": 1.01561332,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 1.780267244985506,
      "language_loss": 0.66690743,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.68884641,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 2.7018723487854004
    },
    {
      "auxiliary_loss_clip": 0.01167403,
      "auxiliary_loss_mlp": 0.0102603,
      "balance_loss_clip": 1.00939822,
      "balance_loss_mlp": 1.01902056,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 1.7396025310192866,
      "language_loss": 0.81967294,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.84160727,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.669987916946411
    },
    {
      "auxiliary_loss_clip": 0.01170579,
      "auxiliary_loss_mlp": 0.01026167,
      "balance_loss_clip": 0.97555381,
      "balance_loss_mlp": 1.01853776,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 8.055686011990947,
      "language_loss": 0.74800247,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.76996994,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.6414635181427
    },
    {
      "auxiliary_loss_clip": 0.01168253,
      "auxiliary_loss_mlp": 0.01023585,
      "balance_loss_clip": 0.97252226,
      "balance_loss_mlp": 1.01620328,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 1.5682938316617943,
      "language_loss": 0.71366346,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.73558187,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 3.7046496868133545
    },
    {
      "auxiliary_loss_clip": 0.01160559,
      "auxiliary_loss_mlp": 0.01023123,
      "balance_loss_clip": 0.89657485,
      "balance_loss_mlp": 1.01546371,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 1.866840210124375,
      "language_loss": 0.80162799,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82346481,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.7463810443878174
    },
    {
      "auxiliary_loss_clip": 0.01161194,
      "auxiliary_loss_mlp": 0.01024394,
      "balance_loss_clip": 0.93155545,
      "balance_loss_mlp": 1.01727152,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.0642977365416,
      "language_loss": 0.77431351,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.7961694,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 3.5707414150238037
    },
    {
      "auxiliary_loss_clip": 0.01167349,
      "auxiliary_loss_mlp": 0.01027112,
      "balance_loss_clip": 1.01201451,
      "balance_loss_mlp": 1.01984334,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.07030593675051,
      "language_loss": 0.67219675,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69414139,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 3.550328016281128
    },
    {
      "auxiliary_loss_clip": 0.0117239,
      "auxiliary_loss_mlp": 0.01027752,
      "balance_loss_clip": 1.05076861,
      "balance_loss_mlp": 1.02035475,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.9716301719258096,
      "language_loss": 0.72031981,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74232125,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.681023597717285
    },
    {
      "auxiliary_loss_clip": 0.01164459,
      "auxiliary_loss_mlp": 0.01023956,
      "balance_loss_clip": 0.9728148,
      "balance_loss_mlp": 1.01639247,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 2.4780087499406456,
      "language_loss": 0.64536703,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.66725123,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.825942277908325
    },
    {
      "auxiliary_loss_clip": 0.01172414,
      "auxiliary_loss_mlp": 0.01034007,
      "balance_loss_clip": 0.9695152,
      "balance_loss_mlp": 1.02615666,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.5761675497946552,
      "language_loss": 0.80806327,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.83012748,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.720932722091675
    },
    {
      "auxiliary_loss_clip": 0.01176635,
      "auxiliary_loss_mlp": 0.01032607,
      "balance_loss_clip": 0.93274021,
      "balance_loss_mlp": 1.02487636,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 1.8969533016861824,
      "language_loss": 0.76007956,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.78217202,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 2.6635565757751465
    },
    {
      "auxiliary_loss_clip": 0.01170962,
      "auxiliary_loss_mlp": 0.01020314,
      "balance_loss_clip": 0.97375274,
      "balance_loss_mlp": 1.0126102,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 1.7317810734383405,
      "language_loss": 0.80982816,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.83174098,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.7742912769317627
    },
    {
      "auxiliary_loss_clip": 0.01174996,
      "auxiliary_loss_mlp": 0.01025899,
      "balance_loss_clip": 0.9761073,
      "balance_loss_mlp": 1.01791811,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.7863032806427441,
      "language_loss": 0.73880291,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76081192,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.627657651901245
    },
    {
      "auxiliary_loss_clip": 0.01154588,
      "auxiliary_loss_mlp": 0.01032578,
      "balance_loss_clip": 0.8180083,
      "balance_loss_mlp": 1.02571177,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 1.9745794024038261,
      "language_loss": 0.84007388,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.86194551,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.8944127559661865
    },
    {
      "auxiliary_loss_clip": 0.01164733,
      "auxiliary_loss_mlp": 0.01024316,
      "balance_loss_clip": 1.01176572,
      "balance_loss_mlp": 1.01644206,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 2.5554194332708047,
      "language_loss": 0.70463097,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.72652149,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.932400703430176
    },
    {
      "auxiliary_loss_clip": 0.01177525,
      "auxiliary_loss_mlp": 0.0103468,
      "balance_loss_clip": 1.05241108,
      "balance_loss_mlp": 1.02712262,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 2.34688168613981,
      "language_loss": 0.7441743,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76629639,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.614301919937134
    },
    {
      "auxiliary_loss_clip": 0.01174546,
      "auxiliary_loss_mlp": 0.01031034,
      "balance_loss_clip": 1.01458192,
      "balance_loss_mlp": 1.02338052,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 1.6057319966573838,
      "language_loss": 0.73123705,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.7532928,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 2.6554653644561768
    },
    {
      "auxiliary_loss_clip": 0.01080742,
      "auxiliary_loss_mlp": 0.01002212,
      "balance_loss_clip": 0.90330267,
      "balance_loss_mlp": 1.00063848,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.7085747051400689,
      "language_loss": 0.57220399,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59303355,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 3.340287208557129
    },
    {
      "auxiliary_loss_clip": 0.01161018,
      "auxiliary_loss_mlp": 0.01024555,
      "balance_loss_clip": 1.00880849,
      "balance_loss_mlp": 1.01782608,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 1.8533995414243307,
      "language_loss": 0.58971477,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.61157048,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.7346839904785156
    },
    {
      "auxiliary_loss_clip": 0.01173529,
      "auxiliary_loss_mlp": 0.01029558,
      "balance_loss_clip": 1.05118418,
      "balance_loss_mlp": 1.02262318,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 1.6950794051511135,
      "language_loss": 0.80098653,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82301742,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.5769805908203125
    },
    {
      "auxiliary_loss_clip": 0.01166836,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.01217103,
      "balance_loss_mlp": 1.02101469,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 1.676080168416026,
      "language_loss": 0.80709296,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.82903987,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 2.6575307846069336
    },
    {
      "auxiliary_loss_clip": 0.0116905,
      "auxiliary_loss_mlp": 0.01026414,
      "balance_loss_clip": 0.97338074,
      "balance_loss_mlp": 1.01889825,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.4531342037736965,
      "language_loss": 0.75210607,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77406073,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.6348581314086914
    },
    {
      "auxiliary_loss_clip": 0.01170616,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 0.97205937,
      "balance_loss_mlp": 1.01865566,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 2.474190693664044,
      "language_loss": 0.81245947,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.83442605,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.675504446029663
    },
    {
      "auxiliary_loss_clip": 0.01071447,
      "auxiliary_loss_mlp": 0.01003027,
      "balance_loss_clip": 1.01629782,
      "balance_loss_mlp": 1.00147688,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 0.849443690305279,
      "language_loss": 0.60438883,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62513363,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 3.1347122192382812
    },
    {
      "auxiliary_loss_clip": 0.01178929,
      "auxiliary_loss_mlp": 0.01029357,
      "balance_loss_clip": 1.01220953,
      "balance_loss_mlp": 1.02176976,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 1.7293009555753784,
      "language_loss": 0.77518493,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79726779,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 3.577697515487671
    },
    {
      "auxiliary_loss_clip": 0.01165727,
      "auxiliary_loss_mlp": 0.01026861,
      "balance_loss_clip": 1.00930417,
      "balance_loss_mlp": 1.01956868,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.4670191064121383,
      "language_loss": 0.753353,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77527893,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 2.7339212894439697
    },
    {
      "auxiliary_loss_clip": 0.01164876,
      "auxiliary_loss_mlp": 0.01027603,
      "balance_loss_clip": 0.93479145,
      "balance_loss_mlp": 1.02057576,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 1.72591243324512,
      "language_loss": 0.74744111,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.76936591,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 2.711637020111084
    },
    {
      "auxiliary_loss_clip": 0.0117965,
      "auxiliary_loss_mlp": 0.01025918,
      "balance_loss_clip": 0.97340947,
      "balance_loss_mlp": 1.01800263,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 5.478442248608229,
      "language_loss": 0.82083428,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.8428899,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.649688482284546
    },
    {
      "auxiliary_loss_clip": 0.01167359,
      "auxiliary_loss_mlp": 0.01024673,
      "balance_loss_clip": 0.97380042,
      "balance_loss_mlp": 1.01703453,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 2.216927418540359,
      "language_loss": 0.6897912,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71171153,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.6834630966186523
    },
    {
      "auxiliary_loss_clip": 0.01175023,
      "auxiliary_loss_mlp": 0.01123061,
      "balance_loss_clip": 0.97675955,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 1.8798127378967786,
      "language_loss": 0.78712791,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.81010878,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 3.632122039794922
    },
    {
      "auxiliary_loss_clip": 0.0116627,
      "auxiliary_loss_mlp": 0.0102465,
      "balance_loss_clip": 0.93728173,
      "balance_loss_mlp": 1.01731229,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 1.8581789913572349,
      "language_loss": 0.73807704,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.75998628,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.682756185531616
    },
    {
      "auxiliary_loss_clip": 0.01173113,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 1.01337874,
      "balance_loss_mlp": 1.02050459,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.017850611513808,
      "language_loss": 0.84494936,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.86695975,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 3.464848756790161
    },
    {
      "auxiliary_loss_clip": 0.0107547,
      "auxiliary_loss_mlp": 0.01003213,
      "balance_loss_clip": 0.9424926,
      "balance_loss_mlp": 1.00155628,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.705862048756631,
      "language_loss": 0.52294767,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54373455,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 4.116872072219849
    },
    {
      "auxiliary_loss_clip": 0.01176446,
      "auxiliary_loss_mlp": 0.01032098,
      "balance_loss_clip": 1.05070722,
      "balance_loss_mlp": 1.0242542,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 1.737670848427465,
      "language_loss": 0.64854348,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67062891,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.808302879333496
    },
    {
      "auxiliary_loss_clip": 0.01172087,
      "auxiliary_loss_mlp": 0.01027122,
      "balance_loss_clip": 0.9368304,
      "balance_loss_mlp": 1.019925,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 1.9568922039698655,
      "language_loss": 0.84661013,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.86860228,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.687039852142334
    },
    {
      "auxiliary_loss_clip": 0.01166392,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 0.92915148,
      "balance_loss_mlp": 1.02079511,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 1.6465495821799563,
      "language_loss": 0.77616298,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.79810995,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.750751256942749
    },
    {
      "auxiliary_loss_clip": 0.01174422,
      "auxiliary_loss_mlp": 0.01024736,
      "balance_loss_clip": 1.01095843,
      "balance_loss_mlp": 1.01738703,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 1.9432449296879821,
      "language_loss": 0.76541525,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.7874068,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 2.650904417037964
    },
    {
      "auxiliary_loss_clip": 0.01167864,
      "auxiliary_loss_mlp": 0.01029998,
      "balance_loss_clip": 0.96910727,
      "balance_loss_mlp": 1.02275562,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.6353509979476735,
      "language_loss": 0.7139734,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73595202,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.788424253463745
    },
    {
      "auxiliary_loss_clip": 0.01172753,
      "auxiliary_loss_mlp": 0.01025828,
      "balance_loss_clip": 0.97646534,
      "balance_loss_mlp": 1.01821899,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 3.002496371830209,
      "language_loss": 0.7209481,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74293399,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.7852282524108887
    },
    {
      "auxiliary_loss_clip": 0.01165526,
      "auxiliary_loss_mlp": 0.01023533,
      "balance_loss_clip": 0.97214341,
      "balance_loss_mlp": 1.01641583,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 1.8128738379079061,
      "language_loss": 0.72650123,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.74839181,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.683102607727051
    },
    {
      "auxiliary_loss_clip": 0.01166284,
      "auxiliary_loss_mlp": 0.01028222,
      "balance_loss_clip": 0.93446851,
      "balance_loss_mlp": 1.02086425,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.6560908931572396,
      "language_loss": 0.80318344,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.82512856,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.729807138442993
    },
    {
      "auxiliary_loss_clip": 0.01177743,
      "auxiliary_loss_mlp": 0.01024182,
      "balance_loss_clip": 1.04999506,
      "balance_loss_mlp": 1.01599848,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 2.1640185273919026,
      "language_loss": 0.80857897,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.83059824,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.6812102794647217
    },
    {
      "auxiliary_loss_clip": 0.01159243,
      "auxiliary_loss_mlp": 0.01024492,
      "balance_loss_clip": 0.89676511,
      "balance_loss_mlp": 1.01725042,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 1.9737355183919922,
      "language_loss": 0.81779051,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.8396278,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.7128593921661377
    },
    {
      "auxiliary_loss_clip": 0.01162452,
      "auxiliary_loss_mlp": 0.01026595,
      "balance_loss_clip": 0.9689135,
      "balance_loss_mlp": 1.01909041,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 1.9167178257671231,
      "language_loss": 0.91457993,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93647039,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 2.6505870819091797
    },
    {
      "auxiliary_loss_clip": 0.01170352,
      "auxiliary_loss_mlp": 0.01022328,
      "balance_loss_clip": 0.97356093,
      "balance_loss_mlp": 1.01530099,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.6497111101107609,
      "language_loss": 0.77111697,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79304379,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.8194589614868164
    },
    {
      "auxiliary_loss_clip": 0.0117584,
      "auxiliary_loss_mlp": 0.01028321,
      "balance_loss_clip": 0.97560066,
      "balance_loss_mlp": 1.02078056,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 2.034114683027051,
      "language_loss": 0.78543472,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.80747628,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 2.704791784286499
    },
    {
      "auxiliary_loss_clip": 0.01176778,
      "auxiliary_loss_mlp": 0.01025347,
      "balance_loss_clip": 0.93603933,
      "balance_loss_mlp": 1.0182004,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.626275943673385,
      "language_loss": 0.76257873,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.78460002,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 2.8719987869262695
    },
    {
      "auxiliary_loss_clip": 0.01171901,
      "auxiliary_loss_mlp": 0.01029543,
      "balance_loss_clip": 1.05227661,
      "balance_loss_mlp": 1.02265906,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.8775463910977963,
      "language_loss": 0.79281181,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81482625,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.6518213748931885
    },
    {
      "auxiliary_loss_clip": 0.01071171,
      "auxiliary_loss_mlp": 0.01002288,
      "balance_loss_clip": 1.0159781,
      "balance_loss_mlp": 1.00075042,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.7713499601648359,
      "language_loss": 0.54438949,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56512409,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.2608680725097656
    },
    {
      "auxiliary_loss_clip": 0.01166782,
      "auxiliary_loss_mlp": 0.01028541,
      "balance_loss_clip": 0.93194389,
      "balance_loss_mlp": 1.02137685,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 2.165821407695865,
      "language_loss": 0.72030163,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74225485,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.7386386394500732
    },
    {
      "auxiliary_loss_clip": 0.01169535,
      "auxiliary_loss_mlp": 0.01032024,
      "balance_loss_clip": 1.05062532,
      "balance_loss_mlp": 1.02502966,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.630467595528921,
      "language_loss": 0.71198726,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73400277,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 3.7575998306274414
    },
    {
      "auxiliary_loss_clip": 0.01172805,
      "auxiliary_loss_mlp": 0.01122467,
      "balance_loss_clip": 1.05192661,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 1.924944879719053,
      "language_loss": 0.76515067,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.78810334,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 2.5950822830200195
    },
    {
      "auxiliary_loss_clip": 0.01170263,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.01287234,
      "balance_loss_mlp": 1.02136099,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.5582838815943747,
      "language_loss": 0.81298304,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83497643,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 2.6705162525177
    },
    {
      "auxiliary_loss_clip": 0.01157296,
      "auxiliary_loss_mlp": 0.01024553,
      "balance_loss_clip": 0.96586019,
      "balance_loss_mlp": 1.01691771,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 1.811029840254445,
      "language_loss": 0.60333401,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62515259,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 2.691159725189209
    },
    {
      "auxiliary_loss_clip": 0.01151493,
      "auxiliary_loss_mlp": 0.01029242,
      "balance_loss_clip": 0.8569144,
      "balance_loss_mlp": 1.02277827,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 1.5768506350154847,
      "language_loss": 0.76154208,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78334945,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.823188543319702
    },
    {
      "auxiliary_loss_clip": 0.01165639,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.010481,
      "balance_loss_mlp": 1.01897252,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.3399635816959847,
      "language_loss": 0.66119164,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.68311715,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 3.5868661403656006
    },
    {
      "auxiliary_loss_clip": 0.01175633,
      "auxiliary_loss_mlp": 0.01025449,
      "balance_loss_clip": 0.9351368,
      "balance_loss_mlp": 1.01804006,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 3.3141558898436623,
      "language_loss": 0.68172938,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.70374024,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 3.6418583393096924
    },
    {
      "auxiliary_loss_clip": 0.01168786,
      "auxiliary_loss_mlp": 0.01028822,
      "balance_loss_clip": 0.97265929,
      "balance_loss_mlp": 1.02064443,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 2.7654363602436454,
      "language_loss": 0.73907351,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.76104963,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.687229871749878
    },
    {
      "auxiliary_loss_clip": 0.01170343,
      "auxiliary_loss_mlp": 0.01023561,
      "balance_loss_clip": 0.85482359,
      "balance_loss_mlp": 1.01622379,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 3.0584560667929654,
      "language_loss": 0.70797724,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.72991627,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 3.639247417449951
    },
    {
      "auxiliary_loss_clip": 0.01172834,
      "auxiliary_loss_mlp": 0.01027824,
      "balance_loss_clip": 0.8923896,
      "balance_loss_mlp": 1.01986074,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.272663159482606,
      "language_loss": 0.75869226,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.7806989,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.7968199253082275
    },
    {
      "auxiliary_loss_clip": 0.01159417,
      "auxiliary_loss_mlp": 0.01029334,
      "balance_loss_clip": 0.96963936,
      "balance_loss_mlp": 1.0215168,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 1.642842211994211,
      "language_loss": 0.78543973,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80732727,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.65165638923645
    },
    {
      "auxiliary_loss_clip": 0.01170302,
      "auxiliary_loss_mlp": 0.01024347,
      "balance_loss_clip": 0.78434467,
      "balance_loss_mlp": 1.01730454,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 1.991415126471734,
      "language_loss": 0.69467485,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.7166214,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.9461796283721924
    },
    {
      "auxiliary_loss_clip": 0.01175246,
      "auxiliary_loss_mlp": 0.01028736,
      "balance_loss_clip": 1.01320982,
      "balance_loss_mlp": 1.02107716,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 2.1910060725591345,
      "language_loss": 0.64779472,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.66983449,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 3.135246992111206
    },
    {
      "auxiliary_loss_clip": 0.0117323,
      "auxiliary_loss_mlp": 0.01023873,
      "balance_loss_clip": 1.00982499,
      "balance_loss_mlp": 1.01663733,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 1.726098699883536,
      "language_loss": 0.77618879,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.79815984,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 2.972799301147461
    },
    {
      "auxiliary_loss_clip": 0.01164334,
      "auxiliary_loss_mlp": 0.01026699,
      "balance_loss_clip": 1.01121497,
      "balance_loss_mlp": 1.01996946,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 1.631944674111674,
      "language_loss": 0.8199991,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.84190941,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.656184434890747
    },
    {
      "auxiliary_loss_clip": 0.01150599,
      "auxiliary_loss_mlp": 0.01027701,
      "balance_loss_clip": 0.9298147,
      "balance_loss_mlp": 1.02017868,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 1.82087144086173,
      "language_loss": 0.84740937,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.86919236,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.6699166297912598
    },
    {
      "auxiliary_loss_clip": 0.01167076,
      "auxiliary_loss_mlp": 0.010252,
      "balance_loss_clip": 0.89770532,
      "balance_loss_mlp": 1.01841426,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 1.7219408704713295,
      "language_loss": 0.8971591,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.91908193,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 3.039466142654419
    },
    {
      "auxiliary_loss_clip": 0.01164564,
      "auxiliary_loss_mlp": 0.01030523,
      "balance_loss_clip": 0.97072721,
      "balance_loss_mlp": 1.02322161,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 1.7445606968100924,
      "language_loss": 0.76218456,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78413546,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.7353928089141846
    },
    {
      "auxiliary_loss_clip": 0.01166878,
      "auxiliary_loss_mlp": 0.01024078,
      "balance_loss_clip": 0.93555206,
      "balance_loss_mlp": 1.01575172,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.0846155286770576,
      "language_loss": 0.75803983,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.77994943,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 2.701449394226074
    },
    {
      "auxiliary_loss_clip": 0.01166646,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 0.93252504,
      "balance_loss_mlp": 1.02155817,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 1.9961746414118517,
      "language_loss": 0.78195083,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.80390871,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 2.641934871673584
    },
    {
      "auxiliary_loss_clip": 0.01165728,
      "auxiliary_loss_mlp": 0.01034667,
      "balance_loss_clip": 0.74051744,
      "balance_loss_mlp": 1.02701092,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 2.079254828649695,
      "language_loss": 0.73604536,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75804931,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 3.1143980026245117
    },
    {
      "auxiliary_loss_clip": 0.01170031,
      "auxiliary_loss_mlp": 0.0103192,
      "balance_loss_clip": 0.97392869,
      "balance_loss_mlp": 1.0245235,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 1.8728475591216966,
      "language_loss": 0.77502632,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.79704589,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 3.367049217224121
    },
    {
      "auxiliary_loss_clip": 0.01169271,
      "auxiliary_loss_mlp": 0.01021429,
      "balance_loss_clip": 1.0098238,
      "balance_loss_mlp": 1.01458669,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.1651256824612353,
      "language_loss": 0.67072189,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69262886,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 2.857494831085205
    },
    {
      "auxiliary_loss_clip": 0.01174846,
      "auxiliary_loss_mlp": 0.01025507,
      "balance_loss_clip": 0.97358274,
      "balance_loss_mlp": 1.01870036,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 1.752323874759085,
      "language_loss": 0.84563243,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86763597,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.7376649379730225
    },
    {
      "auxiliary_loss_clip": 0.01169578,
      "auxiliary_loss_mlp": 0.01122434,
      "balance_loss_clip": 1.00952733,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 2.183752413257876,
      "language_loss": 0.7238555,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.74677563,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 2.6464691162109375
    },
    {
      "auxiliary_loss_clip": 0.01176742,
      "auxiliary_loss_mlp": 0.01022379,
      "balance_loss_clip": 1.0510788,
      "balance_loss_mlp": 1.01479101,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 2.338211267898916,
      "language_loss": 0.75972116,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78171241,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.6657612323760986
    },
    {
      "auxiliary_loss_clip": 0.01179346,
      "auxiliary_loss_mlp": 0.01023713,
      "balance_loss_clip": 0.97603422,
      "balance_loss_mlp": 1.01539469,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 2.6368979635633996,
      "language_loss": 0.6878202,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.70985079,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 3.7463462352752686
    },
    {
      "auxiliary_loss_clip": 0.01160106,
      "auxiliary_loss_mlp": 0.01019593,
      "balance_loss_clip": 0.89356959,
      "balance_loss_mlp": 1.0123446,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 2.1852823828504047,
      "language_loss": 0.80799526,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.82979226,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 2.8339481353759766
    },
    {
      "auxiliary_loss_clip": 0.0115833,
      "auxiliary_loss_mlp": 0.01029827,
      "balance_loss_clip": 0.93337172,
      "balance_loss_mlp": 1.02277851,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 2.3571907032384476,
      "language_loss": 0.7458955,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.76777703,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 2.7719979286193848
    },
    {
      "auxiliary_loss_clip": 0.01180037,
      "auxiliary_loss_mlp": 0.01026313,
      "balance_loss_clip": 0.93794858,
      "balance_loss_mlp": 1.01897311,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 1.7514855323650538,
      "language_loss": 0.74098849,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.76305199,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 2.6814632415771484
    },
    {
      "auxiliary_loss_clip": 0.01176705,
      "auxiliary_loss_mlp": 0.01027823,
      "balance_loss_clip": 1.05122066,
      "balance_loss_mlp": 1.02038419,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 1.828896359861146,
      "language_loss": 0.76523817,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.78728342,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.6479105949401855
    },
    {
      "auxiliary_loss_clip": 0.01163415,
      "auxiliary_loss_mlp": 0.01024805,
      "balance_loss_clip": 0.97041112,
      "balance_loss_mlp": 1.01757538,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.1904700826821104,
      "language_loss": 0.79085886,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.8127411,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.694195032119751
    },
    {
      "auxiliary_loss_clip": 0.01168129,
      "auxiliary_loss_mlp": 0.01028847,
      "balance_loss_clip": 0.97519553,
      "balance_loss_mlp": 1.02163744,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 1.9255413011725402,
      "language_loss": 0.77840859,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80037832,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 3.5993220806121826
    },
    {
      "auxiliary_loss_clip": 0.01176584,
      "auxiliary_loss_mlp": 0.01028228,
      "balance_loss_clip": 1.01639867,
      "balance_loss_mlp": 1.0205152,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.5145651755903164,
      "language_loss": 0.71907401,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.74112213,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 3.66827392578125
    },
    {
      "auxiliary_loss_clip": 0.01068602,
      "auxiliary_loss_mlp": 0.01002575,
      "balance_loss_clip": 0.78835177,
      "balance_loss_mlp": 1.00081086,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7519861666163631,
      "language_loss": 0.52853268,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.5492444,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.4559614658355713
    },
    {
      "auxiliary_loss_clip": 0.01173871,
      "auxiliary_loss_mlp": 0.01027594,
      "balance_loss_clip": 0.93575126,
      "balance_loss_mlp": 1.02040017,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 1.70583762845816,
      "language_loss": 0.79227245,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81428713,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 3.4161670207977295
    },
    {
      "auxiliary_loss_clip": 0.01173627,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 0.93789661,
      "balance_loss_mlp": 1.02194667,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 1.8321537320692867,
      "language_loss": 0.73146749,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.75349343,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 2.7296841144561768
    },
    {
      "auxiliary_loss_clip": 0.01172889,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.04982376,
      "balance_loss_mlp": 1.01881921,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 3.2581558878710757,
      "language_loss": 0.62468338,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.64667499,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.6289923191070557
    },
    {
      "auxiliary_loss_clip": 0.01172949,
      "auxiliary_loss_mlp": 0.01025754,
      "balance_loss_clip": 1.01325762,
      "balance_loss_mlp": 1.01898313,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 1.6912114610887485,
      "language_loss": 0.83145887,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.85344589,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.753183126449585
    },
    {
      "auxiliary_loss_clip": 0.01176762,
      "auxiliary_loss_mlp": 0.01032812,
      "balance_loss_clip": 1.01454663,
      "balance_loss_mlp": 1.02549243,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.2817518202979543,
      "language_loss": 0.8088032,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.830899,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.6374611854553223
    },
    {
      "auxiliary_loss_clip": 0.01168991,
      "auxiliary_loss_mlp": 0.01023864,
      "balance_loss_clip": 0.93140465,
      "balance_loss_mlp": 1.01643765,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 1.7989070056112886,
      "language_loss": 0.72120649,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74313504,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.7119176387786865
    },
    {
      "auxiliary_loss_clip": 0.01167183,
      "auxiliary_loss_mlp": 0.01028922,
      "balance_loss_clip": 0.97229135,
      "balance_loss_mlp": 1.02131641,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 2.1984979076979716,
      "language_loss": 0.86607242,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.88803351,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.6325979232788086
    },
    {
      "auxiliary_loss_clip": 0.01170996,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.0150156,
      "balance_loss_mlp": 1.02356327,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.462617358201191,
      "language_loss": 0.80190521,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82392156,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.6940865516662598
    },
    {
      "auxiliary_loss_clip": 0.0117863,
      "auxiliary_loss_mlp": 0.01024963,
      "balance_loss_clip": 1.01312828,
      "balance_loss_mlp": 1.01757169,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 2.0172921729190345,
      "language_loss": 0.71166444,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.73370034,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.662877082824707
    },
    {
      "auxiliary_loss_clip": 0.01178243,
      "auxiliary_loss_mlp": 0.01122838,
      "balance_loss_clip": 0.93641031,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 1.7944117049639337,
      "language_loss": 0.72497129,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.74798208,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.7366411685943604
    },
    {
      "auxiliary_loss_clip": 0.01077108,
      "auxiliary_loss_mlp": 0.01002989,
      "balance_loss_clip": 0.94153893,
      "balance_loss_mlp": 1.00139177,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.9697718305704279,
      "language_loss": 0.63432181,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65512288,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 3.2715649604797363
    },
    {
      "auxiliary_loss_clip": 0.01174587,
      "auxiliary_loss_mlp": 0.0102641,
      "balance_loss_clip": 1.01258373,
      "balance_loss_mlp": 1.01916218,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 1.7636614395080068,
      "language_loss": 0.74356896,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76557899,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.598201274871826
    },
    {
      "auxiliary_loss_clip": 0.01171917,
      "auxiliary_loss_mlp": 0.01025796,
      "balance_loss_clip": 1.01069903,
      "balance_loss_mlp": 1.01884651,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.4713076522822617,
      "language_loss": 0.72599483,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.74797201,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.6697285175323486
    },
    {
      "auxiliary_loss_clip": 0.01171154,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.01374209,
      "balance_loss_mlp": 1.02314281,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 2.286316253600412,
      "language_loss": 0.78612429,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.8081398,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 2.6210954189300537
    },
    {
      "auxiliary_loss_clip": 0.01170462,
      "auxiliary_loss_mlp": 0.0102655,
      "balance_loss_clip": 1.01123035,
      "balance_loss_mlp": 1.01917148,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 1.7096987208354546,
      "language_loss": 0.78567779,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.80764788,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.7607998847961426
    },
    {
      "auxiliary_loss_clip": 0.01174973,
      "auxiliary_loss_mlp": 0.0102565,
      "balance_loss_clip": 1.01379848,
      "balance_loss_mlp": 1.01815498,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.4710086635831854,
      "language_loss": 0.63655674,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.65856302,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.619511842727661
    },
    {
      "auxiliary_loss_clip": 0.01169728,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 0.93574882,
      "balance_loss_mlp": 1.02239382,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 1.9010091193896534,
      "language_loss": 0.78964543,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.81164122,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 2.6764628887176514
    },
    {
      "auxiliary_loss_clip": 0.01166252,
      "auxiliary_loss_mlp": 0.01024927,
      "balance_loss_clip": 0.89204061,
      "balance_loss_mlp": 1.01701713,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.2140629032736645,
      "language_loss": 0.73505247,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.75696427,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.7351558208465576
    },
    {
      "auxiliary_loss_clip": 0.0116885,
      "auxiliary_loss_mlp": 0.01026112,
      "balance_loss_clip": 1.01201165,
      "balance_loss_mlp": 1.01928473,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 2.021506926951658,
      "language_loss": 0.71442801,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.7363776,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 3.5400912761688232
    },
    {
      "auxiliary_loss_clip": 0.01153693,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 0.85501337,
      "balance_loss_mlp": 1.01831818,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 2.0961163708460946,
      "language_loss": 0.74707109,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.76887059,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 2.821512222290039
    },
    {
      "auxiliary_loss_clip": 0.01173143,
      "auxiliary_loss_mlp": 0.01021186,
      "balance_loss_clip": 1.05051422,
      "balance_loss_mlp": 1.01407862,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 2.618796676609036,
      "language_loss": 0.79918414,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.82112747,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 2.594169855117798
    },
    {
      "auxiliary_loss_clip": 0.0116089,
      "auxiliary_loss_mlp": 0.01030137,
      "balance_loss_clip": 0.93762702,
      "balance_loss_mlp": 1.02287686,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 1.9920107462070997,
      "language_loss": 0.74301034,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76492059,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.7292633056640625
    },
    {
      "auxiliary_loss_clip": 0.01164415,
      "auxiliary_loss_mlp": 0.01021906,
      "balance_loss_clip": 0.89774317,
      "balance_loss_mlp": 1.01483703,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 1.748774626408074,
      "language_loss": 0.78908515,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.81094837,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.8810391426086426
    },
    {
      "auxiliary_loss_clip": 0.0115863,
      "auxiliary_loss_mlp": 0.01026398,
      "balance_loss_clip": 0.85602498,
      "balance_loss_mlp": 1.01934052,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 1.703466761064731,
      "language_loss": 0.80208248,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82393283,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 3.766639232635498
    },
    {
      "auxiliary_loss_clip": 0.01169469,
      "auxiliary_loss_mlp": 0.01029679,
      "balance_loss_clip": 1.01450467,
      "balance_loss_mlp": 1.02271152,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.464951393432187,
      "language_loss": 0.75878406,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.78077543,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.6960084438323975
    },
    {
      "auxiliary_loss_clip": 0.01174647,
      "auxiliary_loss_mlp": 0.01022556,
      "balance_loss_clip": 1.01150155,
      "balance_loss_mlp": 1.01526952,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.1604936374626917,
      "language_loss": 0.70632219,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72829419,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 3.67322039604187
    },
    {
      "auxiliary_loss_clip": 0.0117078,
      "auxiliary_loss_mlp": 0.01028314,
      "balance_loss_clip": 0.89388859,
      "balance_loss_mlp": 1.02081573,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 1.9465282230449399,
      "language_loss": 0.65767032,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.67966127,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.704739809036255
    },
    {
      "auxiliary_loss_clip": 0.01173425,
      "auxiliary_loss_mlp": 0.01025983,
      "balance_loss_clip": 1.01347625,
      "balance_loss_mlp": 1.01898575,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 1.7219787618707616,
      "language_loss": 0.81555575,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.83754981,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.7030014991760254
    },
    {
      "auxiliary_loss_clip": 0.01159562,
      "auxiliary_loss_mlp": 0.01027212,
      "balance_loss_clip": 0.97231036,
      "balance_loss_mlp": 1.01966023,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.0491827831783676,
      "language_loss": 0.77976072,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.80162847,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 2.608365058898926
    },
    {
      "auxiliary_loss_clip": 0.01175031,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 1.01262772,
      "balance_loss_mlp": 1.01970744,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 1.6795811759567023,
      "language_loss": 0.73034906,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75236881,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.6843931674957275
    },
    {
      "auxiliary_loss_clip": 0.01163104,
      "auxiliary_loss_mlp": 0.01024084,
      "balance_loss_clip": 0.93219912,
      "balance_loss_mlp": 1.01668751,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 2.3883471949256347,
      "language_loss": 0.8411665,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86303836,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.6993846893310547
    },
    {
      "auxiliary_loss_clip": 0.01162034,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 0.93157923,
      "balance_loss_mlp": 1.01797724,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 1.989056097207865,
      "language_loss": 0.81306517,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.834934,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.724975109100342
    },
    {
      "auxiliary_loss_clip": 0.01175489,
      "auxiliary_loss_mlp": 0.01025053,
      "balance_loss_clip": 1.01425457,
      "balance_loss_mlp": 1.01836205,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 2.180003331411685,
      "language_loss": 0.82554221,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.84754759,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.6166269779205322
    },
    {
      "auxiliary_loss_clip": 0.01074964,
      "auxiliary_loss_mlp": 0.01000416,
      "balance_loss_clip": 0.86699879,
      "balance_loss_mlp": 0.99871176,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.789367582932735,
      "language_loss": 0.5711844,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59193814,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.4260647296905518
    },
    {
      "auxiliary_loss_clip": 0.01173121,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.01409364,
      "balance_loss_mlp": 1.01873136,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.5616119772057493,
      "language_loss": 0.69429857,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.71629363,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.807809829711914
    },
    {
      "auxiliary_loss_clip": 0.01169991,
      "auxiliary_loss_mlp": 0.0112291,
      "balance_loss_clip": 1.012218,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 2.041301780002135,
      "language_loss": 0.74326921,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.76619822,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.607635021209717
    },
    {
      "auxiliary_loss_clip": 0.01171269,
      "auxiliary_loss_mlp": 0.01122342,
      "balance_loss_clip": 1.05019498,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 1.7305588301728394,
      "language_loss": 0.74379784,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.76673388,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 2.591027021408081
    },
    {
      "auxiliary_loss_clip": 0.01144999,
      "auxiliary_loss_mlp": 0.01022832,
      "balance_loss_clip": 0.85457349,
      "balance_loss_mlp": 1.0153904,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.4872347643103332,
      "language_loss": 0.81226087,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83393919,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.7830872535705566
    },
    {
      "auxiliary_loss_clip": 0.0117203,
      "auxiliary_loss_mlp": 0.01024719,
      "balance_loss_clip": 1.01430631,
      "balance_loss_mlp": 1.01739335,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 2.7058585282239354,
      "language_loss": 0.73688561,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75885308,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 2.6135146617889404
    },
    {
      "auxiliary_loss_clip": 0.0116285,
      "auxiliary_loss_mlp": 0.01022602,
      "balance_loss_clip": 0.96955335,
      "balance_loss_mlp": 1.01528275,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.5370648282170822,
      "language_loss": 0.69732845,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.71918309,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.8435423374176025
    },
    {
      "auxiliary_loss_clip": 0.01160043,
      "auxiliary_loss_mlp": 0.01020277,
      "balance_loss_clip": 0.89045489,
      "balance_loss_mlp": 1.0130167,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.5196626323023479,
      "language_loss": 0.83012921,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85193235,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 2.817350149154663
    },
    {
      "auxiliary_loss_clip": 0.01175286,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 0.97274041,
      "balance_loss_mlp": 1.01843417,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 1.9991402743002882,
      "language_loss": 0.7806946,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.80269992,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.6220901012420654
    },
    {
      "auxiliary_loss_clip": 0.01171207,
      "auxiliary_loss_mlp": 0.0102452,
      "balance_loss_clip": 1.04956341,
      "balance_loss_mlp": 1.01732922,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 1.8488503270185528,
      "language_loss": 0.76312071,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78507793,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 2.6460368633270264
    },
    {
      "auxiliary_loss_clip": 0.01172622,
      "auxiliary_loss_mlp": 0.01122101,
      "balance_loss_clip": 1.01242352,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 1.7914632580014003,
      "language_loss": 0.64842969,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.67137688,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 3.972600221633911
    },
    {
      "auxiliary_loss_clip": 0.01177638,
      "auxiliary_loss_mlp": 0.0102891,
      "balance_loss_clip": 0.97465223,
      "balance_loss_mlp": 1.02101874,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 1.9883516326514679,
      "language_loss": 0.79304183,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81510729,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 2.6789145469665527
    },
    {
      "auxiliary_loss_clip": 0.01174933,
      "auxiliary_loss_mlp": 0.01122695,
      "balance_loss_clip": 0.93496084,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.6115616784176594,
      "language_loss": 0.78202116,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.8049975,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 2.836571216583252
    },
    {
      "auxiliary_loss_clip": 0.01160973,
      "auxiliary_loss_mlp": 0.01026356,
      "balance_loss_clip": 0.89627701,
      "balance_loss_mlp": 1.0186615,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 1.9754615269087485,
      "language_loss": 0.80313772,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82501107,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.710754156112671
    },
    {
      "auxiliary_loss_clip": 0.01155004,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 0.85730469,
      "balance_loss_mlp": 1.0185461,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 1.8901667642126954,
      "language_loss": 0.75408113,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77589035,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 2.8413853645324707
    },
    {
      "auxiliary_loss_clip": 0.01172329,
      "auxiliary_loss_mlp": 0.0102453,
      "balance_loss_clip": 0.97617412,
      "balance_loss_mlp": 1.01730013,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 1.7659184806061539,
      "language_loss": 0.72877645,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.75074506,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.6575794219970703
    },
    {
      "auxiliary_loss_clip": 0.01162279,
      "auxiliary_loss_mlp": 0.01025475,
      "balance_loss_clip": 0.97140509,
      "balance_loss_mlp": 1.01857603,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 1.7137976122132756,
      "language_loss": 0.77211607,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79399359,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 4.484054803848267
    },
    {
      "auxiliary_loss_clip": 0.01172157,
      "auxiliary_loss_mlp": 0.01027347,
      "balance_loss_clip": 1.01469731,
      "balance_loss_mlp": 1.01989079,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 1.9606652290082365,
      "language_loss": 0.81627595,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.83827102,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 3.7850754261016846
    },
    {
      "auxiliary_loss_clip": 0.01154517,
      "auxiliary_loss_mlp": 0.01026723,
      "balance_loss_clip": 0.93249869,
      "balance_loss_mlp": 1.01915574,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 2.138940487978071,
      "language_loss": 0.77066267,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79247504,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.692643404006958
    },
    {
      "auxiliary_loss_clip": 0.01169806,
      "auxiliary_loss_mlp": 0.01023472,
      "balance_loss_clip": 0.97287947,
      "balance_loss_mlp": 1.01660585,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 1.4477100761217385,
      "language_loss": 0.73401022,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.755943,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.6916863918304443
    },
    {
      "auxiliary_loss_clip": 0.0117551,
      "auxiliary_loss_mlp": 0.01028983,
      "balance_loss_clip": 0.97956258,
      "balance_loss_mlp": 1.02200913,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 1.6288902011990776,
      "language_loss": 0.76552987,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78757483,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 2.707145929336548
    },
    {
      "auxiliary_loss_clip": 0.01175154,
      "auxiliary_loss_mlp": 0.01026913,
      "balance_loss_clip": 1.01158476,
      "balance_loss_mlp": 1.01911652,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 1.9114648110926435,
      "language_loss": 0.71070468,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.73272538,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 2.665707588195801
    },
    {
      "auxiliary_loss_clip": 0.01160338,
      "auxiliary_loss_mlp": 0.01029812,
      "balance_loss_clip": 0.93130946,
      "balance_loss_mlp": 1.02294302,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.39048139470583,
      "language_loss": 0.89293444,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.91483593,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.7989859580993652
    },
    {
      "auxiliary_loss_clip": 0.01166132,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.00883222,
      "balance_loss_mlp": 1.02325988,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 2.328654029418337,
      "language_loss": 0.75061095,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77257508,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 2.621924638748169
    },
    {
      "auxiliary_loss_clip": 0.01174246,
      "auxiliary_loss_mlp": 0.01024684,
      "balance_loss_clip": 0.97204149,
      "balance_loss_mlp": 1.01722193,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.6164608724093215,
      "language_loss": 0.74285334,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.76484263,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.8059422969818115
    },
    {
      "auxiliary_loss_clip": 0.01165729,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 0.93632799,
      "balance_loss_mlp": 1.01852202,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 1.7764208769822185,
      "language_loss": 0.70653301,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.72844952,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.8363513946533203
    },
    {
      "auxiliary_loss_clip": 0.01172194,
      "auxiliary_loss_mlp": 0.0103234,
      "balance_loss_clip": 0.97131598,
      "balance_loss_mlp": 1.02468407,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.2406636652825767,
      "language_loss": 0.78230166,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80434692,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.6602602005004883
    },
    {
      "auxiliary_loss_clip": 0.0117555,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 0.97481847,
      "balance_loss_mlp": 1.0168519,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.8495930499580457,
      "language_loss": 0.77953458,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.80153489,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.699272871017456
    },
    {
      "auxiliary_loss_clip": 0.01172055,
      "auxiliary_loss_mlp": 0.0102825,
      "balance_loss_clip": 1.04938591,
      "balance_loss_mlp": 1.02102566,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 1.7541442647660197,
      "language_loss": 0.74803424,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.77003729,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.65234637260437
    },
    {
      "auxiliary_loss_clip": 0.01171,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.01205528,
      "balance_loss_mlp": 1.02023435,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 2.1920096342878055,
      "language_loss": 0.78711337,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.80909491,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 2.6178135871887207
    },
    {
      "auxiliary_loss_clip": 0.01173853,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.01225662,
      "balance_loss_mlp": 1.02037513,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 1.732985098240545,
      "language_loss": 0.76161379,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78363031,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.645230770111084
    },
    {
      "auxiliary_loss_clip": 0.01164304,
      "auxiliary_loss_mlp": 0.01030436,
      "balance_loss_clip": 0.97373009,
      "balance_loss_mlp": 1.02285099,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 3.7522438839079713,
      "language_loss": 0.84691793,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.86886537,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 2.6699676513671875
    },
    {
      "auxiliary_loss_clip": 0.01171054,
      "auxiliary_loss_mlp": 0.01122782,
      "balance_loss_clip": 1.04819489,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.7213659149192395,
      "language_loss": 0.84231418,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.86525249,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.5926156044006348
    },
    {
      "auxiliary_loss_clip": 0.01165505,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 0.93299109,
      "balance_loss_mlp": 1.0189743,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.3480320726095227,
      "language_loss": 0.80223572,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82415664,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.6987009048461914
    },
    {
      "auxiliary_loss_clip": 0.01161997,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 0.97332072,
      "balance_loss_mlp": 1.02222157,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 1.9538810816725891,
      "language_loss": 0.83021021,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85212326,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.6432275772094727
    },
    {
      "auxiliary_loss_clip": 0.01171218,
      "auxiliary_loss_mlp": 0.01024409,
      "balance_loss_clip": 0.97094524,
      "balance_loss_mlp": 1.01724112,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 1.7183113266479226,
      "language_loss": 0.76457924,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.7865355,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 2.7040369510650635
    },
    {
      "auxiliary_loss_clip": 0.01071474,
      "auxiliary_loss_mlp": 0.01000385,
      "balance_loss_clip": 0.97593951,
      "balance_loss_mlp": 0.99883479,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8669054650016347,
      "language_loss": 0.57184005,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59255862,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.1978421211242676
    },
    {
      "auxiliary_loss_clip": 0.01172088,
      "auxiliary_loss_mlp": 0.01023475,
      "balance_loss_clip": 1.01240826,
      "balance_loss_mlp": 1.0163672,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 1.9243187556197952,
      "language_loss": 0.75593156,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77788717,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 3.5313098430633545
    },
    {
      "auxiliary_loss_clip": 0.0116346,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 0.93144441,
      "balance_loss_mlp": 1.01754308,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 1.709349058233637,
      "language_loss": 0.83885849,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.86073935,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 2.667088747024536
    },
    {
      "auxiliary_loss_clip": 0.01163039,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 0.9706918,
      "balance_loss_mlp": 1.01923668,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.2166920859778028,
      "language_loss": 0.8269515,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.84884799,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.6366162300109863
    },
    {
      "auxiliary_loss_clip": 0.01165605,
      "auxiliary_loss_mlp": 0.01025108,
      "balance_loss_clip": 0.97120607,
      "balance_loss_mlp": 1.01751149,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 3.701291022801738,
      "language_loss": 0.89994472,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.92185187,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 2.6864845752716064
    },
    {
      "auxiliary_loss_clip": 0.01164248,
      "auxiliary_loss_mlp": 0.01025348,
      "balance_loss_clip": 0.93434513,
      "balance_loss_mlp": 1.01870501,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.5628534430321226,
      "language_loss": 0.83363229,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85552824,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.6890430450439453
    },
    {
      "auxiliary_loss_clip": 0.01168021,
      "auxiliary_loss_mlp": 0.01024303,
      "balance_loss_clip": 1.01204062,
      "balance_loss_mlp": 1.01694143,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 3.7472597917312567,
      "language_loss": 0.74825382,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.77017701,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 5.40869402885437
    },
    {
      "auxiliary_loss_clip": 0.01167686,
      "auxiliary_loss_mlp": 0.01027745,
      "balance_loss_clip": 1.01196742,
      "balance_loss_mlp": 1.02100396,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 1.72147066533808,
      "language_loss": 0.72620404,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74815834,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.6605114936828613
    },
    {
      "auxiliary_loss_clip": 0.01170199,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 1.01276934,
      "balance_loss_mlp": 1.01742077,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 1.8829294258082276,
      "language_loss": 0.77036643,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.79232216,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.6338560581207275
    },
    {
      "auxiliary_loss_clip": 0.01172081,
      "auxiliary_loss_mlp": 0.01023845,
      "balance_loss_clip": 1.05067587,
      "balance_loss_mlp": 1.01656175,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.958535830421578,
      "language_loss": 0.75485456,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.77681386,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.588498830795288
    },
    {
      "auxiliary_loss_clip": 0.01071189,
      "auxiliary_loss_mlp": 0.01002844,
      "balance_loss_clip": 0.97636008,
      "balance_loss_mlp": 1.00135982,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.8099738513056588,
      "language_loss": 0.57914639,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59988666,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.0682859420776367
    },
    {
      "auxiliary_loss_clip": 0.01164338,
      "auxiliary_loss_mlp": 0.01024971,
      "balance_loss_clip": 0.97095931,
      "balance_loss_mlp": 1.01734757,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 1.931432126549577,
      "language_loss": 0.64570892,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66760206,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.76190447807312
    },
    {
      "auxiliary_loss_clip": 0.01169108,
      "auxiliary_loss_mlp": 0.01024986,
      "balance_loss_clip": 1.04956269,
      "balance_loss_mlp": 1.01805067,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 1.7319262884734488,
      "language_loss": 0.72305685,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.7449978,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.5645852088928223
    },
    {
      "auxiliary_loss_clip": 0.01169303,
      "auxiliary_loss_mlp": 0.01123077,
      "balance_loss_clip": 0.90105391,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.6130949397794678,
      "language_loss": 0.78190351,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.80482733,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.8011410236358643
    },
    {
      "auxiliary_loss_clip": 0.01175017,
      "auxiliary_loss_mlp": 0.0102829,
      "balance_loss_clip": 0.93730378,
      "balance_loss_mlp": 1.02027607,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 1.905083071702668,
      "language_loss": 0.82007283,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84210587,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.6721320152282715
    },
    {
      "auxiliary_loss_clip": 0.01171938,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 1.01246691,
      "balance_loss_mlp": 1.01926064,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.3245621836177923,
      "language_loss": 0.83895522,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.86094266,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.6136181354522705
    },
    {
      "auxiliary_loss_clip": 0.01176372,
      "auxiliary_loss_mlp": 0.01025058,
      "balance_loss_clip": 0.97470969,
      "balance_loss_mlp": 1.01751828,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 1.7146071352464882,
      "language_loss": 0.7120111,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73402536,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.656660318374634
    },
    {
      "auxiliary_loss_clip": 0.01170671,
      "auxiliary_loss_mlp": 0.01025276,
      "balance_loss_clip": 1.01139069,
      "balance_loss_mlp": 1.01783454,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 1.7636708727270223,
      "language_loss": 0.81978655,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.84174597,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.6563332080841064
    },
    {
      "auxiliary_loss_clip": 0.01172377,
      "auxiliary_loss_mlp": 0.0102429,
      "balance_loss_clip": 1.01277101,
      "balance_loss_mlp": 1.01761413,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 1.7411181446654405,
      "language_loss": 0.81127375,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.83324045,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.7495532035827637
    },
    {
      "auxiliary_loss_clip": 0.01166599,
      "auxiliary_loss_mlp": 0.01024906,
      "balance_loss_clip": 1.0120585,
      "balance_loss_mlp": 1.01770282,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 1.9067666020655958,
      "language_loss": 0.761944,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78385907,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 2.7334561347961426
    },
    {
      "auxiliary_loss_clip": 0.01169492,
      "auxiliary_loss_mlp": 0.0102315,
      "balance_loss_clip": 1.05019855,
      "balance_loss_mlp": 1.01607823,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 1.6548610631226042,
      "language_loss": 0.7425189,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.7644453,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.654085636138916
    },
    {
      "auxiliary_loss_clip": 0.0117126,
      "auxiliary_loss_mlp": 0.01024813,
      "balance_loss_clip": 1.01281428,
      "balance_loss_mlp": 1.0171566,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 1.700686648592255,
      "language_loss": 0.69946039,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.72142112,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 2.6607913970947266
    },
    {
      "auxiliary_loss_clip": 0.0117939,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 0.97517753,
      "balance_loss_mlp": 1.02086473,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 2.2130973910011753,
      "language_loss": 0.83155149,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.85363257,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 2.7471306324005127
    },
    {
      "auxiliary_loss_clip": 0.01167418,
      "auxiliary_loss_mlp": 0.01024928,
      "balance_loss_clip": 0.93244857,
      "balance_loss_mlp": 1.01777804,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 1.861240135038781,
      "language_loss": 0.75867915,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.78060257,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.7757740020751953
    },
    {
      "auxiliary_loss_clip": 0.01171602,
      "auxiliary_loss_mlp": 0.01027751,
      "balance_loss_clip": 1.01338577,
      "balance_loss_mlp": 1.02090812,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.5090208242317293,
      "language_loss": 0.77870524,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80069882,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 2.675257444381714
    },
    {
      "auxiliary_loss_clip": 0.0117387,
      "auxiliary_loss_mlp": 0.01027677,
      "balance_loss_clip": 1.04857254,
      "balance_loss_mlp": 1.02061105,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 1.9222090968023184,
      "language_loss": 0.67674923,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.69876468,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.590062141418457
    },
    {
      "auxiliary_loss_clip": 0.01173679,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 0.93570518,
      "balance_loss_mlp": 1.01869571,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 1.6304594314171683,
      "language_loss": 0.80888712,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.83088714,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 2.6795167922973633
    },
    {
      "auxiliary_loss_clip": 0.01071933,
      "auxiliary_loss_mlp": 0.0099885,
      "balance_loss_clip": 0.90002048,
      "balance_loss_mlp": 0.99721634,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.734842157648244,
      "language_loss": 0.54313278,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.56384057,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 4.12629508972168
    },
    {
      "auxiliary_loss_clip": 0.01169496,
      "auxiliary_loss_mlp": 0.01028972,
      "balance_loss_clip": 0.97568393,
      "balance_loss_mlp": 1.02170622,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.5599952667506403,
      "language_loss": 0.83724809,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.85923278,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 2.621541738510132
    },
    {
      "auxiliary_loss_clip": 0.01158317,
      "auxiliary_loss_mlp": 0.01022385,
      "balance_loss_clip": 0.97149444,
      "balance_loss_mlp": 1.01504445,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 1.7399624693541775,
      "language_loss": 0.7224853,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.74429238,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.6269524097442627
    },
    {
      "auxiliary_loss_clip": 0.01170327,
      "auxiliary_loss_mlp": 0.01025611,
      "balance_loss_clip": 0.97125435,
      "balance_loss_mlp": 1.01861978,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 2.600858546168585,
      "language_loss": 0.75317204,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.7751314,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.661386013031006
    },
    {
      "auxiliary_loss_clip": 0.01168753,
      "auxiliary_loss_mlp": 0.01029213,
      "balance_loss_clip": 0.93725282,
      "balance_loss_mlp": 1.02164936,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.0333430276615503,
      "language_loss": 0.70777428,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.72975397,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 3.58663010597229
    },
    {
      "auxiliary_loss_clip": 0.01169815,
      "auxiliary_loss_mlp": 0.01023832,
      "balance_loss_clip": 1.00984645,
      "balance_loss_mlp": 1.01703095,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 1.5387074944256782,
      "language_loss": 0.76168478,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78362119,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 3.6257457733154297
    },
    {
      "auxiliary_loss_clip": 0.01076157,
      "auxiliary_loss_mlp": 0.01003647,
      "balance_loss_clip": 0.87023103,
      "balance_loss_mlp": 1.00188243,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.822714934231164,
      "language_loss": 0.59001201,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.61081004,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.2988667488098145
    },
    {
      "auxiliary_loss_clip": 0.01174148,
      "auxiliary_loss_mlp": 0.01027513,
      "balance_loss_clip": 0.93382663,
      "balance_loss_mlp": 1.01968122,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 1.6991513217744783,
      "language_loss": 0.70150685,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72352338,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 2.7289695739746094
    },
    {
      "auxiliary_loss_clip": 0.01166679,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 0.89117229,
      "balance_loss_mlp": 1.01987398,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.7266006880741216,
      "language_loss": 0.76740956,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.78934443,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.8237202167510986
    },
    {
      "auxiliary_loss_clip": 0.01174808,
      "auxiliary_loss_mlp": 0.01025918,
      "balance_loss_clip": 0.97260678,
      "balance_loss_mlp": 1.01858652,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 1.5989325873821627,
      "language_loss": 0.76592481,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.78793204,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.7505762577056885
    },
    {
      "auxiliary_loss_clip": 0.01169722,
      "auxiliary_loss_mlp": 0.01023333,
      "balance_loss_clip": 1.01037264,
      "balance_loss_mlp": 1.01595998,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.6368894238808485,
      "language_loss": 0.70384163,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.72577214,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 2.8815131187438965
    },
    {
      "auxiliary_loss_clip": 0.01173261,
      "auxiliary_loss_mlp": 0.01025448,
      "balance_loss_clip": 0.97383124,
      "balance_loss_mlp": 1.01848626,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 1.7914137970714992,
      "language_loss": 0.73424363,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75623071,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.6992411613464355
    },
    {
      "auxiliary_loss_clip": 0.01176711,
      "auxiliary_loss_mlp": 0.01025212,
      "balance_loss_clip": 1.0537765,
      "balance_loss_mlp": 1.01771998,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 2.4674384485150767,
      "language_loss": 0.77513826,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.79715753,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.654125452041626
    },
    {
      "auxiliary_loss_clip": 0.01161635,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 0.97068387,
      "balance_loss_mlp": 1.01903582,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 1.6184152499475353,
      "language_loss": 0.7481603,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.77004451,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.719010353088379
    },
    {
      "auxiliary_loss_clip": 0.01171167,
      "auxiliary_loss_mlp": 0.01025003,
      "balance_loss_clip": 1.01183128,
      "balance_loss_mlp": 1.01750481,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 1.9809244455480681,
      "language_loss": 0.70966434,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.73162603,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.684224843978882
    },
    {
      "auxiliary_loss_clip": 0.01170268,
      "auxiliary_loss_mlp": 0.01026859,
      "balance_loss_clip": 1.01125956,
      "balance_loss_mlp": 1.01922703,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 1.820281038041891,
      "language_loss": 0.80902672,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.83099806,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.6858296394348145
    },
    {
      "auxiliary_loss_clip": 0.01081573,
      "auxiliary_loss_mlp": 0.01001156,
      "balance_loss_clip": 0.95110208,
      "balance_loss_mlp": 0.99932063,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7844471320688458,
      "language_loss": 0.53410184,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55492914,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.2464888095855713
    },
    {
      "auxiliary_loss_clip": 0.01175895,
      "auxiliary_loss_mlp": 0.01026344,
      "balance_loss_clip": 1.01317632,
      "balance_loss_mlp": 1.01823211,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 2.121032520841048,
      "language_loss": 0.76657391,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.78859627,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.6792054176330566
    },
    {
      "auxiliary_loss_clip": 0.0117256,
      "auxiliary_loss_mlp": 0.01025147,
      "balance_loss_clip": 0.97519225,
      "balance_loss_mlp": 1.01716626,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 1.8628304895343542,
      "language_loss": 0.75643957,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.77841669,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 2.7274653911590576
    },
    {
      "auxiliary_loss_clip": 0.01172649,
      "auxiliary_loss_mlp": 0.01123015,
      "balance_loss_clip": 1.05006719,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 2.212839920416578,
      "language_loss": 0.72585452,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.74881113,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.7712299823760986
    },
    {
      "auxiliary_loss_clip": 0.01164734,
      "auxiliary_loss_mlp": 0.01029049,
      "balance_loss_clip": 0.89241958,
      "balance_loss_mlp": 1.02144909,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 2.1505943651086152,
      "language_loss": 0.63629967,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.65823746,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 2.698920726776123
    },
    {
      "auxiliary_loss_clip": 0.01173617,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 0.93853295,
      "balance_loss_mlp": 1.02155626,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.5777454138239402,
      "language_loss": 0.74325407,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.76528227,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.7440147399902344
    },
    {
      "auxiliary_loss_clip": 0.01172067,
      "auxiliary_loss_mlp": 0.01026945,
      "balance_loss_clip": 0.97453368,
      "balance_loss_mlp": 1.01938152,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 3.759883226061594,
      "language_loss": 0.6159035,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.63789356,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.8082034587860107
    },
    {
      "auxiliary_loss_clip": 0.01175794,
      "auxiliary_loss_mlp": 0.01027061,
      "balance_loss_clip": 1.05023837,
      "balance_loss_mlp": 1.01860023,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 3.1664911005173155,
      "language_loss": 0.8215462,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.84357476,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 2.718379259109497
    },
    {
      "auxiliary_loss_clip": 0.01175316,
      "auxiliary_loss_mlp": 0.01026102,
      "balance_loss_clip": 1.0111022,
      "balance_loss_mlp": 1.01857364,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 2.0688503454860303,
      "language_loss": 0.79422206,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81623626,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.650423526763916
    },
    {
      "auxiliary_loss_clip": 0.01168752,
      "auxiliary_loss_mlp": 0.01031082,
      "balance_loss_clip": 0.97128487,
      "balance_loss_mlp": 1.02318406,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 2.255501636088057,
      "language_loss": 0.77711785,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.79911613,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.669645071029663
    },
    {
      "auxiliary_loss_clip": 0.01166483,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 0.96929228,
      "balance_loss_mlp": 1.02383161,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 1.7333664028914717,
      "language_loss": 0.72466511,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.74664408,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 3.613328456878662
    },
    {
      "auxiliary_loss_clip": 0.01169892,
      "auxiliary_loss_mlp": 0.01027476,
      "balance_loss_clip": 0.97132349,
      "balance_loss_mlp": 1.0196383,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 4.694282963657878,
      "language_loss": 0.75948644,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.78146005,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 2.6465823650360107
    },
    {
      "auxiliary_loss_clip": 0.01171286,
      "auxiliary_loss_mlp": 0.01026163,
      "balance_loss_clip": 1.01326847,
      "balance_loss_mlp": 1.0186646,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 1.7229041256363633,
      "language_loss": 0.75578576,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.77776027,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.6876542568206787
    },
    {
      "auxiliary_loss_clip": 0.01164735,
      "auxiliary_loss_mlp": 0.01029467,
      "balance_loss_clip": 0.93506938,
      "balance_loss_mlp": 1.02164674,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 2.3068984920026554,
      "language_loss": 0.70626497,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.72820699,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.6847572326660156
    },
    {
      "auxiliary_loss_clip": 0.0117137,
      "auxiliary_loss_mlp": 0.01027799,
      "balance_loss_clip": 0.9735142,
      "balance_loss_mlp": 1.02013946,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.3144654608798882,
      "language_loss": 0.88333666,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90532827,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 3.538104772567749
    },
    {
      "auxiliary_loss_clip": 0.01170404,
      "auxiliary_loss_mlp": 0.01021793,
      "balance_loss_clip": 1.04758286,
      "balance_loss_mlp": 1.01482511,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 1.6219092129847557,
      "language_loss": 0.7716009,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.79352289,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 3.5684525966644287
    },
    {
      "auxiliary_loss_clip": 0.0116064,
      "auxiliary_loss_mlp": 0.01122401,
      "balance_loss_clip": 0.8940503,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 1.8724381648478776,
      "language_loss": 0.75993896,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.78276938,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 3.6893129348754883
    },
    {
      "auxiliary_loss_clip": 0.01172579,
      "auxiliary_loss_mlp": 0.0102534,
      "balance_loss_clip": 1.01134753,
      "balance_loss_mlp": 1.0181334,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 2.1726918366851464,
      "language_loss": 0.74216145,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76414061,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.749373197555542
    },
    {
      "auxiliary_loss_clip": 0.01172134,
      "auxiliary_loss_mlp": 0.01023112,
      "balance_loss_clip": 1.01331747,
      "balance_loss_mlp": 1.01551843,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.3527962412280474,
      "language_loss": 0.64047641,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.66242886,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 2.743912696838379
    },
    {
      "auxiliary_loss_clip": 0.01162615,
      "auxiliary_loss_mlp": 0.01024411,
      "balance_loss_clip": 1.00903893,
      "balance_loss_mlp": 1.01709127,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 1.5341395801518165,
      "language_loss": 0.71898687,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.74085718,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 2.7438321113586426
    },
    {
      "auxiliary_loss_clip": 0.01171161,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.04896879,
      "balance_loss_mlp": 1.02147293,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.3163780882456986,
      "language_loss": 0.63580596,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.65780592,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.592410087585449
    },
    {
      "auxiliary_loss_clip": 0.01170835,
      "auxiliary_loss_mlp": 0.01121481,
      "balance_loss_clip": 1.01076102,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 5.200143532158661,
      "language_loss": 0.77789015,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.80081332,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 2.659801959991455
    },
    {
      "auxiliary_loss_clip": 0.01167624,
      "auxiliary_loss_mlp": 0.01024258,
      "balance_loss_clip": 0.9739086,
      "balance_loss_mlp": 1.01716161,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 1.5920776750864367,
      "language_loss": 0.81334436,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83526319,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.789611339569092
    },
    {
      "auxiliary_loss_clip": 0.01162349,
      "auxiliary_loss_mlp": 0.01024395,
      "balance_loss_clip": 1.00827265,
      "balance_loss_mlp": 1.0171175,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 1.7158931574813165,
      "language_loss": 0.72859466,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75046206,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.714845895767212
    },
    {
      "auxiliary_loss_clip": 0.01169537,
      "auxiliary_loss_mlp": 0.01022183,
      "balance_loss_clip": 0.89476693,
      "balance_loss_mlp": 1.01562035,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 1.7267656110135898,
      "language_loss": 0.80849636,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.83041358,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.874546527862549
    },
    {
      "auxiliary_loss_clip": 0.01161399,
      "auxiliary_loss_mlp": 0.01024151,
      "balance_loss_clip": 1.00867724,
      "balance_loss_mlp": 1.01699829,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.5966530970160002,
      "language_loss": 0.7737838,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79563928,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.7047407627105713
    },
    {
      "auxiliary_loss_clip": 0.01172977,
      "auxiliary_loss_mlp": 0.01025879,
      "balance_loss_clip": 0.936818,
      "balance_loss_mlp": 1.01818681,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 4.553297918317413,
      "language_loss": 0.73619497,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.7581836,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.768402099609375
    },
    {
      "auxiliary_loss_clip": 0.01174996,
      "auxiliary_loss_mlp": 0.01027711,
      "balance_loss_clip": 1.01064467,
      "balance_loss_mlp": 1.01960158,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 1.9864677545543021,
      "language_loss": 0.72377557,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.74580264,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 2.601083517074585
    },
    {
      "auxiliary_loss_clip": 0.011587,
      "auxiliary_loss_mlp": 0.01021952,
      "balance_loss_clip": 0.89169735,
      "balance_loss_mlp": 1.01359558,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 2.4052434622854943,
      "language_loss": 0.6180191,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.6398257,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 2.949490547180176
    },
    {
      "auxiliary_loss_clip": 0.01170013,
      "auxiliary_loss_mlp": 0.01025531,
      "balance_loss_clip": 1.01196527,
      "balance_loss_mlp": 1.01818752,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 1.4551131729603048,
      "language_loss": 0.73027098,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75222647,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.6650590896606445
    },
    {
      "auxiliary_loss_clip": 0.0117615,
      "auxiliary_loss_mlp": 0.01024106,
      "balance_loss_clip": 1.05178738,
      "balance_loss_mlp": 1.01694155,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 3.2447186742071046,
      "language_loss": 0.71927053,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.74127311,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 2.6049468517303467
    },
    {
      "auxiliary_loss_clip": 0.0116741,
      "auxiliary_loss_mlp": 0.01025099,
      "balance_loss_clip": 1.01141834,
      "balance_loss_mlp": 1.01859355,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 1.6278413575168253,
      "language_loss": 0.69000423,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71192932,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 2.6421163082122803
    },
    {
      "auxiliary_loss_clip": 0.01152986,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 0.85535467,
      "balance_loss_mlp": 1.02155495,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 1.6876251259031023,
      "language_loss": 0.84749514,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.86931235,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.695808172225952
    },
    {
      "auxiliary_loss_clip": 0.01172812,
      "auxiliary_loss_mlp": 0.01025136,
      "balance_loss_clip": 1.05009508,
      "balance_loss_mlp": 1.01739895,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 4.240540402238748,
      "language_loss": 0.82277608,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84475553,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.592836618423462
    },
    {
      "auxiliary_loss_clip": 0.01165718,
      "auxiliary_loss_mlp": 0.01023312,
      "balance_loss_clip": 0.9342376,
      "balance_loss_mlp": 1.0160135,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.6642340251331496,
      "language_loss": 0.72801626,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.74990654,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 2.779064178466797
    },
    {
      "auxiliary_loss_clip": 0.01151715,
      "auxiliary_loss_mlp": 0.01023745,
      "balance_loss_clip": 0.93124288,
      "balance_loss_mlp": 1.01677752,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 1.8519080208710812,
      "language_loss": 0.69087791,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.71263254,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.8404152393341064
    },
    {
      "auxiliary_loss_clip": 0.01168767,
      "auxiliary_loss_mlp": 0.0103015,
      "balance_loss_clip": 0.8977015,
      "balance_loss_mlp": 1.02316463,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 3.236198399780482,
      "language_loss": 0.80013311,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.82212228,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 3.6185688972473145
    },
    {
      "auxiliary_loss_clip": 0.01172608,
      "auxiliary_loss_mlp": 0.01028431,
      "balance_loss_clip": 0.97213024,
      "balance_loss_mlp": 1.02119517,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 2.361278453348682,
      "language_loss": 0.78601974,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.80803013,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 2.6563563346862793
    },
    {
      "auxiliary_loss_clip": 0.01161225,
      "auxiliary_loss_mlp": 0.01023123,
      "balance_loss_clip": 0.92937356,
      "balance_loss_mlp": 1.01562476,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 8.771617093052908,
      "language_loss": 0.75729686,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.77914035,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 2.7170538902282715
    },
    {
      "auxiliary_loss_clip": 0.01170858,
      "auxiliary_loss_mlp": 0.0102827,
      "balance_loss_clip": 1.01331115,
      "balance_loss_mlp": 1.02107263,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 2.144535741140699,
      "language_loss": 0.83252937,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.85452068,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 2.6907451152801514
    },
    {
      "auxiliary_loss_clip": 0.01166866,
      "auxiliary_loss_mlp": 0.01030566,
      "balance_loss_clip": 1.01024759,
      "balance_loss_mlp": 1.02370524,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 2.4096572755206567,
      "language_loss": 0.73737061,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.75934494,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 3.5594398975372314
    },
    {
      "auxiliary_loss_clip": 0.01158207,
      "auxiliary_loss_mlp": 0.0102656,
      "balance_loss_clip": 0.96911561,
      "balance_loss_mlp": 1.01920223,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 1.41636433195908,
      "language_loss": 0.79919517,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82104284,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 3.671154260635376
    },
    {
      "auxiliary_loss_clip": 0.01168392,
      "auxiliary_loss_mlp": 0.01030445,
      "balance_loss_clip": 0.97011065,
      "balance_loss_mlp": 1.02305436,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 1.8897705142658854,
      "language_loss": 0.8560608,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.87804925,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 3.6555635929107666
    },
    {
      "auxiliary_loss_clip": 0.01160613,
      "auxiliary_loss_mlp": 0.01026694,
      "balance_loss_clip": 0.93299127,
      "balance_loss_mlp": 1.01928496,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 1.5711904083456711,
      "language_loss": 0.78495526,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.80682832,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.7447638511657715
    },
    {
      "auxiliary_loss_clip": 0.0116238,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 0.93044239,
      "balance_loss_mlp": 1.01806998,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 4.732942397776752,
      "language_loss": 0.79523325,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81711078,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.800708293914795
    },
    {
      "auxiliary_loss_clip": 0.01165584,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.01159263,
      "balance_loss_mlp": 1.02010632,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 1.823036859416767,
      "language_loss": 0.82000399,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.84193653,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 2.628899097442627
    },
    {
      "auxiliary_loss_clip": 0.01156886,
      "auxiliary_loss_mlp": 0.01123048,
      "balance_loss_clip": 0.89421797,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 1.8353560528756376,
      "language_loss": 0.77656829,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.79936761,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.7805004119873047
    },
    {
      "auxiliary_loss_clip": 0.01157936,
      "auxiliary_loss_mlp": 0.01122649,
      "balance_loss_clip": 0.85163355,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 3.723133693756934,
      "language_loss": 0.634758,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65756381,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.814318895339966
    },
    {
      "auxiliary_loss_clip": 0.01163371,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 0.97260499,
      "balance_loss_mlp": 1.01958382,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.6371097862012314,
      "language_loss": 0.67939532,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.70129573,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 2.7947373390197754
    },
    {
      "auxiliary_loss_clip": 0.01171074,
      "auxiliary_loss_mlp": 0.01030371,
      "balance_loss_clip": 0.89360243,
      "balance_loss_mlp": 1.02296829,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 1.9222371982762194,
      "language_loss": 0.60547948,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.62749398,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.8291492462158203
    },
    {
      "auxiliary_loss_clip": 0.01172087,
      "auxiliary_loss_mlp": 0.01026098,
      "balance_loss_clip": 1.01321197,
      "balance_loss_mlp": 1.0179913,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 1.5747591577434619,
      "language_loss": 0.74925864,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77124053,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.6677582263946533
    },
    {
      "auxiliary_loss_clip": 0.01164396,
      "auxiliary_loss_mlp": 0.01027336,
      "balance_loss_clip": 0.85660404,
      "balance_loss_mlp": 1.02001321,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 1.6210798405446418,
      "language_loss": 0.7532959,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77521324,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 2.763575315475464
    },
    {
      "auxiliary_loss_clip": 0.01162778,
      "auxiliary_loss_mlp": 0.01030131,
      "balance_loss_clip": 0.85566258,
      "balance_loss_mlp": 1.02178669,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 2.0774038935286687,
      "language_loss": 0.8118794,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.83380842,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.721489667892456
    },
    {
      "auxiliary_loss_clip": 0.01159639,
      "auxiliary_loss_mlp": 0.01122433,
      "balance_loss_clip": 0.96886182,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 2.2458450780519015,
      "language_loss": 0.78149426,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.80431497,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 2.6435601711273193
    },
    {
      "auxiliary_loss_clip": 0.01171437,
      "auxiliary_loss_mlp": 0.01021382,
      "balance_loss_clip": 0.97354758,
      "balance_loss_mlp": 1.01443172,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 2.2918644667670307,
      "language_loss": 0.79866117,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82058936,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.7809267044067383
    },
    {
      "auxiliary_loss_clip": 0.01170664,
      "auxiliary_loss_mlp": 0.0102378,
      "balance_loss_clip": 1.01143789,
      "balance_loss_mlp": 1.01683605,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 1.9812293207641156,
      "language_loss": 0.7333287,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.7552731,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.729140043258667
    },
    {
      "auxiliary_loss_clip": 0.01174605,
      "auxiliary_loss_mlp": 0.01026,
      "balance_loss_clip": 1.01506972,
      "balance_loss_mlp": 1.01856434,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 1.8234052718864648,
      "language_loss": 0.77414691,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79615301,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 2.715773344039917
    },
    {
      "auxiliary_loss_clip": 0.011706,
      "auxiliary_loss_mlp": 0.01122822,
      "balance_loss_clip": 1.01081157,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 2.4928954094323332,
      "language_loss": 0.74485099,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.76778519,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.624626874923706
    },
    {
      "auxiliary_loss_clip": 0.0116774,
      "auxiliary_loss_mlp": 0.01025118,
      "balance_loss_clip": 0.96930456,
      "balance_loss_mlp": 1.01736951,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 1.5599429484523513,
      "language_loss": 0.82530314,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84723175,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.736445188522339
    },
    {
      "auxiliary_loss_clip": 0.01171689,
      "auxiliary_loss_mlp": 0.01027444,
      "balance_loss_clip": 1.01483083,
      "balance_loss_mlp": 1.02007389,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 2.4923062250001555,
      "language_loss": 0.78029835,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.80228972,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.7767653465270996
    },
    {
      "auxiliary_loss_clip": 0.01175618,
      "auxiliary_loss_mlp": 0.01025837,
      "balance_loss_clip": 1.01266575,
      "balance_loss_mlp": 1.0180999,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 2.262425090850409,
      "language_loss": 0.7496053,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.7716198,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 2.6251273155212402
    },
    {
      "auxiliary_loss_clip": 0.01075145,
      "auxiliary_loss_mlp": 0.01002555,
      "balance_loss_clip": 0.90500712,
      "balance_loss_mlp": 1.00099373,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7362821671904001,
      "language_loss": 0.56359351,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58437049,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.243414878845215
    },
    {
      "auxiliary_loss_clip": 0.01169626,
      "auxiliary_loss_mlp": 0.01025512,
      "balance_loss_clip": 0.97356945,
      "balance_loss_mlp": 1.01803184,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.5936556568351357,
      "language_loss": 0.64465404,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.66660547,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 3.6925859451293945
    },
    {
      "auxiliary_loss_clip": 0.01165571,
      "auxiliary_loss_mlp": 0.01122241,
      "balance_loss_clip": 0.97438526,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 2.8744864807431387,
      "language_loss": 0.7710492,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.79392731,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 2.704146146774292
    },
    {
      "auxiliary_loss_clip": 0.01074514,
      "auxiliary_loss_mlp": 0.01001261,
      "balance_loss_clip": 0.93917781,
      "balance_loss_mlp": 0.99960417,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.8078255454666585,
      "language_loss": 0.57616782,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.59692562,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.211615562438965
    },
    {
      "auxiliary_loss_clip": 0.0116715,
      "auxiliary_loss_mlp": 0.01025203,
      "balance_loss_clip": 0.9714613,
      "balance_loss_mlp": 1.01851869,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 2.7557185149989136,
      "language_loss": 0.90181273,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92373627,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 2.662710428237915
    },
    {
      "auxiliary_loss_clip": 0.01171532,
      "auxiliary_loss_mlp": 0.01027539,
      "balance_loss_clip": 1.01091468,
      "balance_loss_mlp": 1.02014184,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 1.7509368163201349,
      "language_loss": 0.65980059,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68179131,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.624279737472534
    },
    {
      "auxiliary_loss_clip": 0.01169696,
      "auxiliary_loss_mlp": 0.01024497,
      "balance_loss_clip": 1.01003373,
      "balance_loss_mlp": 1.01731181,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 3.649729069956956,
      "language_loss": 0.71595085,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.73789281,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 3.5824289321899414
    },
    {
      "auxiliary_loss_clip": 0.01162844,
      "auxiliary_loss_mlp": 0.01025772,
      "balance_loss_clip": 1.00919211,
      "balance_loss_mlp": 1.01860738,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.6336070102893552,
      "language_loss": 0.75242692,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.77431309,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 3.574777841567993
    },
    {
      "auxiliary_loss_clip": 0.0116057,
      "auxiliary_loss_mlp": 0.01021147,
      "balance_loss_clip": 0.93547773,
      "balance_loss_mlp": 1.01398563,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.4514286375761905,
      "language_loss": 0.6666612,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.68847841,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.8610143661499023
    },
    {
      "auxiliary_loss_clip": 0.01169497,
      "auxiliary_loss_mlp": 0.01029216,
      "balance_loss_clip": 0.97536761,
      "balance_loss_mlp": 1.02220345,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 1.7228843723068927,
      "language_loss": 0.80281329,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82480049,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.6668789386749268
    },
    {
      "auxiliary_loss_clip": 0.01173775,
      "auxiliary_loss_mlp": 0.01030111,
      "balance_loss_clip": 1.05125737,
      "balance_loss_mlp": 1.02285409,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 1.8065051021690302,
      "language_loss": 0.71387911,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73591805,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.5935442447662354
    },
    {
      "auxiliary_loss_clip": 0.01162835,
      "auxiliary_loss_mlp": 0.0102865,
      "balance_loss_clip": 0.93459308,
      "balance_loss_mlp": 1.0214169,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 1.3954347619554253,
      "language_loss": 0.69886178,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72077668,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.696943998336792
    },
    {
      "auxiliary_loss_clip": 0.01168249,
      "auxiliary_loss_mlp": 0.01122912,
      "balance_loss_clip": 1.00862408,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 2.3846343320167698,
      "language_loss": 0.80036151,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.82327318,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.657942533493042
    },
    {
      "auxiliary_loss_clip": 0.0116315,
      "auxiliary_loss_mlp": 0.01026861,
      "balance_loss_clip": 0.93232775,
      "balance_loss_mlp": 1.02005696,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 1.8945220998461982,
      "language_loss": 0.84039664,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.86229682,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 2.670311689376831
    },
    {
      "auxiliary_loss_clip": 0.01172189,
      "auxiliary_loss_mlp": 0.01030328,
      "balance_loss_clip": 0.93523967,
      "balance_loss_mlp": 1.02316391,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.788584841129075,
      "language_loss": 0.70364487,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.72567004,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.6806628704071045
    },
    {
      "auxiliary_loss_clip": 0.01168774,
      "auxiliary_loss_mlp": 0.01022979,
      "balance_loss_clip": 1.01098228,
      "balance_loss_mlp": 1.01539147,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 1.652619581519308,
      "language_loss": 0.69708741,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.71900499,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.6353063583374023
    },
    {
      "auxiliary_loss_clip": 0.01162606,
      "auxiliary_loss_mlp": 0.01024057,
      "balance_loss_clip": 1.00967503,
      "balance_loss_mlp": 1.01728833,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 1.5380634281003978,
      "language_loss": 0.73595053,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.75781721,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.7478110790252686
    },
    {
      "auxiliary_loss_clip": 0.01172158,
      "auxiliary_loss_mlp": 0.01031823,
      "balance_loss_clip": 1.05131137,
      "balance_loss_mlp": 1.02497697,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 2.8699345181803744,
      "language_loss": 0.71535963,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.73739946,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 2.682427167892456
    },
    {
      "auxiliary_loss_clip": 0.01171682,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 0.9711622,
      "balance_loss_mlp": 1.01850224,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 6.790427825880287,
      "language_loss": 0.77899647,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.80097246,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.6949641704559326
    },
    {
      "auxiliary_loss_clip": 0.01165519,
      "auxiliary_loss_mlp": 0.01025122,
      "balance_loss_clip": 0.92976969,
      "balance_loss_mlp": 1.01729941,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 2.1212495397420983,
      "language_loss": 0.76159376,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.78350019,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.70078706741333
    },
    {
      "auxiliary_loss_clip": 0.01164807,
      "auxiliary_loss_mlp": 0.01023469,
      "balance_loss_clip": 0.97342253,
      "balance_loss_mlp": 1.0164088,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 1.876970635067376,
      "language_loss": 0.76245922,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.78434199,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.7756707668304443
    },
    {
      "auxiliary_loss_clip": 0.01156087,
      "auxiliary_loss_mlp": 0.01027311,
      "balance_loss_clip": 0.89288664,
      "balance_loss_mlp": 1.02050376,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 1.8961908312313003,
      "language_loss": 0.81254238,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.83437634,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 2.745086908340454
    },
    {
      "auxiliary_loss_clip": 0.01172378,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 0.97410762,
      "balance_loss_mlp": 1.01896405,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 1.6151522981272095,
      "language_loss": 0.80511117,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82709372,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.6245737075805664
    },
    {
      "auxiliary_loss_clip": 0.01170845,
      "auxiliary_loss_mlp": 0.01023026,
      "balance_loss_clip": 1.00975347,
      "balance_loss_mlp": 1.0160464,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 1.6863744551354816,
      "language_loss": 0.66079241,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68273109,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.691035032272339
    },
    {
      "auxiliary_loss_clip": 0.0115584,
      "auxiliary_loss_mlp": 0.01026482,
      "balance_loss_clip": 0.96864069,
      "balance_loss_mlp": 1.01918066,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 1.8372424082322736,
      "language_loss": 0.75591338,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.77773666,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 2.7053589820861816
    },
    {
      "auxiliary_loss_clip": 0.01174504,
      "auxiliary_loss_mlp": 0.01022447,
      "balance_loss_clip": 1.0510751,
      "balance_loss_mlp": 1.01526523,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 2.4509549026026627,
      "language_loss": 0.72424304,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.74621254,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.6537134647369385
    },
    {
      "auxiliary_loss_clip": 0.01169708,
      "auxiliary_loss_mlp": 0.01021993,
      "balance_loss_clip": 1.01265836,
      "balance_loss_mlp": 1.01498318,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 2.281487765749376,
      "language_loss": 0.75923306,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78115004,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.6664586067199707
    },
    {
      "auxiliary_loss_clip": 0.0117568,
      "auxiliary_loss_mlp": 0.01025712,
      "balance_loss_clip": 1.01196027,
      "balance_loss_mlp": 1.01796985,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 2.1879889980777776,
      "language_loss": 0.79991257,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82192647,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 3.649228096008301
    },
    {
      "auxiliary_loss_clip": 0.01156922,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 0.85974109,
      "balance_loss_mlp": 1.02200294,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 1.9749594537173842,
      "language_loss": 0.72897112,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75083864,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.8653557300567627
    },
    {
      "auxiliary_loss_clip": 0.01155756,
      "auxiliary_loss_mlp": 0.01022392,
      "balance_loss_clip": 0.92814445,
      "balance_loss_mlp": 1.01513243,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 2.519820566319009,
      "language_loss": 0.82015687,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.84193832,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 2.6548051834106445
    },
    {
      "auxiliary_loss_clip": 0.01159161,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 0.93599147,
      "balance_loss_mlp": 1.02206373,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 2.119951475799084,
      "language_loss": 0.77922559,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.80111033,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 2.756286382675171
    },
    {
      "auxiliary_loss_clip": 0.01165946,
      "auxiliary_loss_mlp": 0.01024813,
      "balance_loss_clip": 1.01112247,
      "balance_loss_mlp": 1.01825619,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 1.9844046974557492,
      "language_loss": 0.77670014,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.79860771,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 3.5151138305664062
    },
    {
      "auxiliary_loss_clip": 0.01176075,
      "auxiliary_loss_mlp": 0.0102448,
      "balance_loss_clip": 1.04891801,
      "balance_loss_mlp": 1.01669586,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 2.77729544680056,
      "language_loss": 0.8390305,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.861036,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.6576344966888428
    },
    {
      "auxiliary_loss_clip": 0.01168788,
      "auxiliary_loss_mlp": 0.01027178,
      "balance_loss_clip": 0.93381178,
      "balance_loss_mlp": 1.01940262,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.0168847703312665,
      "language_loss": 0.84796488,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.86992455,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 3.6765296459198
    },
    {
      "auxiliary_loss_clip": 0.01164687,
      "auxiliary_loss_mlp": 0.01024188,
      "balance_loss_clip": 0.96819615,
      "balance_loss_mlp": 1.01674342,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 1.787050707299595,
      "language_loss": 0.82980382,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85169256,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.6686313152313232
    },
    {
      "auxiliary_loss_clip": 0.01168504,
      "auxiliary_loss_mlp": 0.0102812,
      "balance_loss_clip": 1.04985094,
      "balance_loss_mlp": 1.02130723,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 1.5902506531651441,
      "language_loss": 0.79900503,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.82097131,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.6949710845947266
    },
    {
      "auxiliary_loss_clip": 0.01166975,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.04737353,
      "balance_loss_mlp": 1.01880503,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.538351183093456,
      "language_loss": 0.67199177,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69392371,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 2.726693630218506
    },
    {
      "auxiliary_loss_clip": 0.011668,
      "auxiliary_loss_mlp": 0.01028596,
      "balance_loss_clip": 0.8946166,
      "balance_loss_mlp": 1.02044189,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.102580590482813,
      "language_loss": 0.78652394,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80847794,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.733553886413574
    },
    {
      "auxiliary_loss_clip": 0.01169029,
      "auxiliary_loss_mlp": 0.010271,
      "balance_loss_clip": 1.01138139,
      "balance_loss_mlp": 1.01979566,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.2146198927769305,
      "language_loss": 0.76211923,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.78408051,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.7775566577911377
    },
    {
      "auxiliary_loss_clip": 0.01157569,
      "auxiliary_loss_mlp": 0.01026065,
      "balance_loss_clip": 0.97159815,
      "balance_loss_mlp": 1.01919246,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.57467045000349,
      "language_loss": 0.69895995,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72079629,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 2.681152105331421
    },
    {
      "auxiliary_loss_clip": 0.01163374,
      "auxiliary_loss_mlp": 0.01023801,
      "balance_loss_clip": 0.89607096,
      "balance_loss_mlp": 1.01688647,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 1.5271188941180547,
      "language_loss": 0.68103814,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70290995,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.894343376159668
    },
    {
      "auxiliary_loss_clip": 0.0117413,
      "auxiliary_loss_mlp": 0.01025509,
      "balance_loss_clip": 1.01319456,
      "balance_loss_mlp": 1.01798069,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 1.6351845804452196,
      "language_loss": 0.80798703,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.82998347,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.6187775135040283
    },
    {
      "auxiliary_loss_clip": 0.01172807,
      "auxiliary_loss_mlp": 0.01025805,
      "balance_loss_clip": 1.05208826,
      "balance_loss_mlp": 1.0186466,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 1.7912085983892048,
      "language_loss": 0.8109045,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83289057,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.603976249694824
    },
    {
      "auxiliary_loss_clip": 0.01171306,
      "auxiliary_loss_mlp": 0.010257,
      "balance_loss_clip": 1.04850018,
      "balance_loss_mlp": 1.01827955,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 2.0153294951808536,
      "language_loss": 0.73941553,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.76138562,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 2.6443839073181152
    },
    {
      "auxiliary_loss_clip": 0.01170406,
      "auxiliary_loss_mlp": 0.01026091,
      "balance_loss_clip": 1.05126476,
      "balance_loss_mlp": 1.01956105,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 1.8034793355372294,
      "language_loss": 0.76481223,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.78677726,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.598902702331543
    },
    {
      "auxiliary_loss_clip": 0.01081175,
      "auxiliary_loss_mlp": 0.0100062,
      "balance_loss_clip": 0.90315348,
      "balance_loss_mlp": 0.99899834,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.8947496431932915,
      "language_loss": 0.62603676,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64685476,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.2685961723327637
    },
    {
      "auxiliary_loss_clip": 0.01161874,
      "auxiliary_loss_mlp": 0.01033971,
      "balance_loss_clip": 0.9333235,
      "balance_loss_mlp": 1.02598405,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 1.6826346931574603,
      "language_loss": 0.79502082,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.81697929,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.79559588432312
    },
    {
      "auxiliary_loss_clip": 0.01171363,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.04960322,
      "balance_loss_mlp": 1.02001989,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 2.209470752537504,
      "language_loss": 0.7216692,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74365389,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 2.526806592941284
    },
    {
      "auxiliary_loss_clip": 0.0115664,
      "auxiliary_loss_mlp": 0.01032436,
      "balance_loss_clip": 0.78088462,
      "balance_loss_mlp": 1.02520013,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.1659066186572984,
      "language_loss": 0.83791208,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.85980284,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 2.951721668243408
    },
    {
      "auxiliary_loss_clip": 0.01176189,
      "auxiliary_loss_mlp": 0.01031749,
      "balance_loss_clip": 0.93731403,
      "balance_loss_mlp": 1.02377963,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 1.843415359871783,
      "language_loss": 0.75694132,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.77902073,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 3.1554882526397705
    },
    {
      "auxiliary_loss_clip": 0.01153393,
      "auxiliary_loss_mlp": 0.01028808,
      "balance_loss_clip": 0.89622962,
      "balance_loss_mlp": 1.02154565,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 3.813020521304631,
      "language_loss": 0.7511487,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.7729708,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.7216289043426514
    },
    {
      "auxiliary_loss_clip": 0.01169446,
      "auxiliary_loss_mlp": 0.01024609,
      "balance_loss_clip": 1.04867363,
      "balance_loss_mlp": 1.01673281,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 1.6913094004047218,
      "language_loss": 0.78127861,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.8032192,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 2.6581881046295166
    },
    {
      "auxiliary_loss_clip": 0.01067389,
      "auxiliary_loss_mlp": 0.01003135,
      "balance_loss_clip": 1.01316774,
      "balance_loss_mlp": 1.00154936,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.7833568137438208,
      "language_loss": 0.57536948,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.5960747,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.2367868423461914
    },
    {
      "auxiliary_loss_clip": 0.01163909,
      "auxiliary_loss_mlp": 0.01030768,
      "balance_loss_clip": 0.97219074,
      "balance_loss_mlp": 1.02353144,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 1.8515062569373477,
      "language_loss": 0.80936849,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83131528,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 4.054861783981323
    },
    {
      "auxiliary_loss_clip": 0.01174335,
      "auxiliary_loss_mlp": 0.01030326,
      "balance_loss_clip": 1.04999411,
      "balance_loss_mlp": 1.02281857,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 1.826015212458188,
      "language_loss": 0.72225964,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74430621,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.662010908126831
    },
    {
      "auxiliary_loss_clip": 0.01169787,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 0.97363126,
      "balance_loss_mlp": 1.0174551,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.673591327647134,
      "language_loss": 0.71990186,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.74184668,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.721069812774658
    },
    {
      "auxiliary_loss_clip": 0.01163253,
      "auxiliary_loss_mlp": 0.0112235,
      "balance_loss_clip": 0.97051406,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.177736258580397,
      "language_loss": 0.80843103,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.83128703,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.6777658462524414
    },
    {
      "auxiliary_loss_clip": 0.01069175,
      "auxiliary_loss_mlp": 0.0100286,
      "balance_loss_clip": 0.93823355,
      "balance_loss_mlp": 1.0012387,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7801330953608385,
      "language_loss": 0.59804618,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61876655,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.1752676963806152
    },
    {
      "auxiliary_loss_clip": 0.01070544,
      "auxiliary_loss_mlp": 0.01001481,
      "balance_loss_clip": 0.97540426,
      "balance_loss_mlp": 0.99991912,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8484827413959484,
      "language_loss": 0.55121368,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.57193398,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 4.908107042312622
    },
    {
      "auxiliary_loss_clip": 0.01164324,
      "auxiliary_loss_mlp": 0.01030865,
      "balance_loss_clip": 0.97157204,
      "balance_loss_mlp": 1.02400434,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 1.8054039856748192,
      "language_loss": 0.82345927,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.84541118,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 3.592411756515503
    },
    {
      "auxiliary_loss_clip": 0.0116676,
      "auxiliary_loss_mlp": 0.01030535,
      "balance_loss_clip": 1.01297593,
      "balance_loss_mlp": 1.02365649,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 1.6778860350670148,
      "language_loss": 0.83788109,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.85985404,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 2.703653573989868
    },
    {
      "auxiliary_loss_clip": 0.0116896,
      "auxiliary_loss_mlp": 0.01024073,
      "balance_loss_clip": 0.93337202,
      "balance_loss_mlp": 1.01694691,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 1.8916490928800682,
      "language_loss": 0.717116,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.73904634,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.718641757965088
    },
    {
      "auxiliary_loss_clip": 0.01169071,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 0.93450075,
      "balance_loss_mlp": 1.01967812,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 1.7572829449581397,
      "language_loss": 0.7924931,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81445789,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.7631404399871826
    },
    {
      "auxiliary_loss_clip": 0.01171906,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 0.97201729,
      "balance_loss_mlp": 1.01617265,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 2.2499056882879587,
      "language_loss": 0.72833699,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75029135,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.6424973011016846
    },
    {
      "auxiliary_loss_clip": 0.01175147,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 0.97403479,
      "balance_loss_mlp": 1.02446222,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 3.8920034859792723,
      "language_loss": 0.76001859,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78209245,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.64426589012146
    },
    {
      "auxiliary_loss_clip": 0.01166553,
      "auxiliary_loss_mlp": 0.01026566,
      "balance_loss_clip": 0.97435558,
      "balance_loss_mlp": 1.01909173,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 2.1809572545263927,
      "language_loss": 0.73477513,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75670636,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.6729629039764404
    },
    {
      "auxiliary_loss_clip": 0.01167396,
      "auxiliary_loss_mlp": 0.01025627,
      "balance_loss_clip": 0.93330956,
      "balance_loss_mlp": 1.01818788,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 1.6753232690860456,
      "language_loss": 0.54790139,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.56983161,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 2.8143582344055176
    },
    {
      "auxiliary_loss_clip": 0.0116843,
      "auxiliary_loss_mlp": 0.01025894,
      "balance_loss_clip": 1.04810798,
      "balance_loss_mlp": 1.01840794,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 3.768236909649301,
      "language_loss": 0.77294779,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.794891,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.550318956375122
    },
    {
      "auxiliary_loss_clip": 0.01164937,
      "auxiliary_loss_mlp": 0.01020315,
      "balance_loss_clip": 0.97061384,
      "balance_loss_mlp": 1.0134958,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.1462453326011652,
      "language_loss": 0.76427829,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.78613085,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 2.7217888832092285
    },
    {
      "auxiliary_loss_clip": 0.01176672,
      "auxiliary_loss_mlp": 0.01025952,
      "balance_loss_clip": 0.97356784,
      "balance_loss_mlp": 1.01850128,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 2.308251893507947,
      "language_loss": 0.75714135,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.77916765,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 2.6727700233459473
    },
    {
      "auxiliary_loss_clip": 0.01170128,
      "auxiliary_loss_mlp": 0.01022252,
      "balance_loss_clip": 0.97325128,
      "balance_loss_mlp": 1.0150938,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.7583166105514565,
      "language_loss": 0.811526,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.83344978,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 2.764101028442383
    },
    {
      "auxiliary_loss_clip": 0.01167541,
      "auxiliary_loss_mlp": 0.01023788,
      "balance_loss_clip": 0.9708994,
      "balance_loss_mlp": 1.01603913,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.427738166402404,
      "language_loss": 0.73657238,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.75848567,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.689955949783325
    },
    {
      "auxiliary_loss_clip": 0.01165713,
      "auxiliary_loss_mlp": 0.01023589,
      "balance_loss_clip": 0.9312259,
      "balance_loss_mlp": 1.0161978,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 2.404747968499722,
      "language_loss": 0.89493465,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.91682762,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.7242870330810547
    },
    {
      "auxiliary_loss_clip": 0.011728,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 0.89746904,
      "balance_loss_mlp": 1.01912844,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 1.8910006870120095,
      "language_loss": 0.86812031,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.89010602,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 2.7492945194244385
    },
    {
      "auxiliary_loss_clip": 0.0117619,
      "auxiliary_loss_mlp": 0.01034592,
      "balance_loss_clip": 1.0147078,
      "balance_loss_mlp": 1.02734447,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.550445253564321,
      "language_loss": 0.68595684,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.70806468,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.6241390705108643
    },
    {
      "auxiliary_loss_clip": 0.0116869,
      "auxiliary_loss_mlp": 0.01028143,
      "balance_loss_clip": 0.97381818,
      "balance_loss_mlp": 1.02113056,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.0847795877837383,
      "language_loss": 0.78696644,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.80893481,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.697467088699341
    },
    {
      "auxiliary_loss_clip": 0.01167251,
      "auxiliary_loss_mlp": 0.01026414,
      "balance_loss_clip": 0.93505895,
      "balance_loss_mlp": 1.01859355,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.417330642498015,
      "language_loss": 0.79814726,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.82008398,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 2.6920058727264404
    },
    {
      "auxiliary_loss_clip": 0.01170013,
      "auxiliary_loss_mlp": 0.01027546,
      "balance_loss_clip": 0.97159219,
      "balance_loss_mlp": 1.02026868,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.694938251766929,
      "language_loss": 0.83006507,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85204065,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.763140916824341
    },
    {
      "auxiliary_loss_clip": 0.01166712,
      "auxiliary_loss_mlp": 0.01029951,
      "balance_loss_clip": 0.93296945,
      "balance_loss_mlp": 1.02184522,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 3.137263666489519,
      "language_loss": 0.64882898,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.67079568,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 3.655857801437378
    },
    {
      "auxiliary_loss_clip": 0.01067562,
      "auxiliary_loss_mlp": 0.01001029,
      "balance_loss_clip": 0.90437651,
      "balance_loss_mlp": 0.99939537,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7869493416507688,
      "language_loss": 0.56658947,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58727539,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.2186062335968018
    },
    {
      "auxiliary_loss_clip": 0.01175423,
      "auxiliary_loss_mlp": 0.01023388,
      "balance_loss_clip": 1.05076575,
      "balance_loss_mlp": 1.01570463,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 2.4575705766417504,
      "language_loss": 0.86095178,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.88293993,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.6474530696868896
    },
    {
      "auxiliary_loss_clip": 0.01168935,
      "auxiliary_loss_mlp": 0.01024995,
      "balance_loss_clip": 0.9307133,
      "balance_loss_mlp": 1.01765728,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.6390322184385862,
      "language_loss": 0.63797587,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.65991515,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.8130764961242676
    },
    {
      "auxiliary_loss_clip": 0.01164957,
      "auxiliary_loss_mlp": 0.01027136,
      "balance_loss_clip": 1.01053107,
      "balance_loss_mlp": 1.01937246,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.225610527255174,
      "language_loss": 0.70940375,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.73132473,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 2.7333643436431885
    },
    {
      "auxiliary_loss_clip": 0.01162422,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.00940847,
      "balance_loss_mlp": 1.01710224,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 2.265719597527753,
      "language_loss": 0.69647741,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.71834761,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 3.5638058185577393
    },
    {
      "auxiliary_loss_clip": 0.01167122,
      "auxiliary_loss_mlp": 0.01025907,
      "balance_loss_clip": 0.8952046,
      "balance_loss_mlp": 1.01888549,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.4753913004937385,
      "language_loss": 0.75593054,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.77786076,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 3.7062594890594482
    },
    {
      "auxiliary_loss_clip": 0.01169547,
      "auxiliary_loss_mlp": 0.01029035,
      "balance_loss_clip": 0.97579044,
      "balance_loss_mlp": 1.02165639,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 1.6391912423068404,
      "language_loss": 0.75644964,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.77843541,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 3.626051187515259
    },
    {
      "auxiliary_loss_clip": 0.01168725,
      "auxiliary_loss_mlp": 0.01025623,
      "balance_loss_clip": 1.04863942,
      "balance_loss_mlp": 1.01881897,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 1.8675116864301728,
      "language_loss": 0.71148443,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73342788,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.658820629119873
    },
    {
      "auxiliary_loss_clip": 0.01171125,
      "auxiliary_loss_mlp": 0.01025788,
      "balance_loss_clip": 1.00890827,
      "balance_loss_mlp": 1.01836169,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 1.6842336441236878,
      "language_loss": 0.78785729,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.80982643,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 2.6468942165374756
    },
    {
      "auxiliary_loss_clip": 0.01155583,
      "auxiliary_loss_mlp": 0.01024026,
      "balance_loss_clip": 0.93305671,
      "balance_loss_mlp": 1.01734436,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 3.497903582799157,
      "language_loss": 0.78210628,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.80390239,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.680879592895508
    },
    {
      "auxiliary_loss_clip": 0.01165127,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 0.97143143,
      "balance_loss_mlp": 1.01892543,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 1.963001589039381,
      "language_loss": 0.73459387,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.75650334,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.5736477375030518
    },
    {
      "auxiliary_loss_clip": 0.01165359,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 0.78360432,
      "balance_loss_mlp": 1.01709199,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 1.916627635670406,
      "language_loss": 0.62626076,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.64815652,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 3.0087730884552
    },
    {
      "auxiliary_loss_clip": 0.01165427,
      "auxiliary_loss_mlp": 0.01022638,
      "balance_loss_clip": 0.93136173,
      "balance_loss_mlp": 1.01558101,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 2.0415180663644747,
      "language_loss": 0.73997581,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.76185644,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 2.963127613067627
    },
    {
      "auxiliary_loss_clip": 0.01162735,
      "auxiliary_loss_mlp": 0.01028754,
      "balance_loss_clip": 1.00976062,
      "balance_loss_mlp": 1.02136314,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.4542968597138515,
      "language_loss": 0.80876446,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.8306793,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 2.682868003845215
    },
    {
      "auxiliary_loss_clip": 0.01165433,
      "auxiliary_loss_mlp": 0.01023041,
      "balance_loss_clip": 1.00899887,
      "balance_loss_mlp": 1.01498568,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 1.8271939902514522,
      "language_loss": 0.82718801,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.84907281,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.6088340282440186
    },
    {
      "auxiliary_loss_clip": 0.01166068,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 1.01233613,
      "balance_loss_mlp": 1.01622248,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 2.549959347011034,
      "language_loss": 0.64071977,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.66261184,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 2.6262435913085938
    },
    {
      "auxiliary_loss_clip": 0.01067239,
      "auxiliary_loss_mlp": 0.01002693,
      "balance_loss_clip": 1.0130235,
      "balance_loss_mlp": 1.00111961,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.7737345147976794,
      "language_loss": 0.56055963,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.58125889,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 3.0660955905914307
    },
    {
      "auxiliary_loss_clip": 0.01174504,
      "auxiliary_loss_mlp": 0.01029005,
      "balance_loss_clip": 1.05088496,
      "balance_loss_mlp": 1.02152801,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 4.945522214162882,
      "language_loss": 0.71434522,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.73638022,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 2.58292555809021
    },
    {
      "auxiliary_loss_clip": 0.01176499,
      "auxiliary_loss_mlp": 0.01030272,
      "balance_loss_clip": 1.01265597,
      "balance_loss_mlp": 1.02225494,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 2.6316632368111836,
      "language_loss": 0.75806463,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.78013235,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.604980230331421
    },
    {
      "auxiliary_loss_clip": 0.01171769,
      "auxiliary_loss_mlp": 0.01023074,
      "balance_loss_clip": 0.97260273,
      "balance_loss_mlp": 1.01609123,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 1.523957665666801,
      "language_loss": 0.7214964,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.7434448,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 2.7008001804351807
    },
    {
      "auxiliary_loss_clip": 0.01162768,
      "auxiliary_loss_mlp": 0.01024309,
      "balance_loss_clip": 1.01112366,
      "balance_loss_mlp": 1.01741314,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 1.7589687623400208,
      "language_loss": 0.78748339,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.80935419,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 2.6079936027526855
    },
    {
      "auxiliary_loss_clip": 0.01174248,
      "auxiliary_loss_mlp": 0.01027883,
      "balance_loss_clip": 1.01075625,
      "balance_loss_mlp": 1.02010143,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 3.0065981273578277,
      "language_loss": 0.63205075,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.65407205,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.62214994430542
    },
    {
      "auxiliary_loss_clip": 0.01170337,
      "auxiliary_loss_mlp": 0.01028925,
      "balance_loss_clip": 1.01467645,
      "balance_loss_mlp": 1.021415,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 1.7163970652772538,
      "language_loss": 0.69590789,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71790051,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.693668842315674
    },
    {
      "auxiliary_loss_clip": 0.01169293,
      "auxiliary_loss_mlp": 0.01026403,
      "balance_loss_clip": 1.01291871,
      "balance_loss_mlp": 1.01979232,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.7455595037368377,
      "language_loss": 0.72466016,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.74661714,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.638317108154297
    },
    {
      "auxiliary_loss_clip": 0.01167978,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.04828131,
      "balance_loss_mlp": 1.01983225,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 1.8133485496168862,
      "language_loss": 0.63793421,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.65987659,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.6983537673950195
    },
    {
      "auxiliary_loss_clip": 0.01158612,
      "auxiliary_loss_mlp": 0.0102826,
      "balance_loss_clip": 0.97031826,
      "balance_loss_mlp": 1.02066898,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 2.031382067049503,
      "language_loss": 0.75181556,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77368426,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.7136716842651367
    },
    {
      "auxiliary_loss_clip": 0.01173841,
      "auxiliary_loss_mlp": 0.01026085,
      "balance_loss_clip": 0.97560585,
      "balance_loss_mlp": 1.01840782,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 1.5002724895206827,
      "language_loss": 0.67610097,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69810027,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 3.600037097930908
    },
    {
      "auxiliary_loss_clip": 0.01164466,
      "auxiliary_loss_mlp": 0.01027097,
      "balance_loss_clip": 0.93095851,
      "balance_loss_mlp": 1.01994467,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 1.856350082965629,
      "language_loss": 0.7322551,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75417072,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.749659538269043
    },
    {
      "auxiliary_loss_clip": 0.01153681,
      "auxiliary_loss_mlp": 0.0102642,
      "balance_loss_clip": 0.92961276,
      "balance_loss_mlp": 1.01993227,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 2.043330905424155,
      "language_loss": 0.7434299,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76523095,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.7207605838775635
    },
    {
      "auxiliary_loss_clip": 0.01168625,
      "auxiliary_loss_mlp": 0.01025483,
      "balance_loss_clip": 1.01153505,
      "balance_loss_mlp": 1.01794291,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 2.108282071058105,
      "language_loss": 0.77988183,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.8018229,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.714796543121338
    },
    {
      "auxiliary_loss_clip": 0.01175098,
      "auxiliary_loss_mlp": 0.01025223,
      "balance_loss_clip": 0.93640339,
      "balance_loss_mlp": 1.01768911,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 2.329657537333104,
      "language_loss": 0.72085339,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.74285656,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.716700315475464
    },
    {
      "auxiliary_loss_clip": 0.01176567,
      "auxiliary_loss_mlp": 0.01031627,
      "balance_loss_clip": 0.94214356,
      "balance_loss_mlp": 1.0238843,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 1.8929129101400208,
      "language_loss": 0.71533972,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.73742175,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 5.065800189971924
    },
    {
      "auxiliary_loss_clip": 0.01170069,
      "auxiliary_loss_mlp": 0.01028042,
      "balance_loss_clip": 1.04696047,
      "balance_loss_mlp": 1.02022815,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.7539899781872355,
      "language_loss": 0.77057827,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.79255939,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 3.5618488788604736
    },
    {
      "auxiliary_loss_clip": 0.01173923,
      "auxiliary_loss_mlp": 0.01018568,
      "balance_loss_clip": 1.05169261,
      "balance_loss_mlp": 1.01122475,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 2.660438601676744,
      "language_loss": 0.82035637,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84228122,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.564166307449341
    },
    {
      "auxiliary_loss_clip": 0.0117205,
      "auxiliary_loss_mlp": 0.01023787,
      "balance_loss_clip": 1.01076794,
      "balance_loss_mlp": 1.01687264,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.698481782284807,
      "language_loss": 0.74608493,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76804328,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 2.644487142562866
    },
    {
      "auxiliary_loss_clip": 0.01173305,
      "auxiliary_loss_mlp": 0.01122663,
      "balance_loss_clip": 1.04992986,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 2.440766291091994,
      "language_loss": 0.84323573,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86619544,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 2.5704774856567383
    },
    {
      "auxiliary_loss_clip": 0.01174201,
      "auxiliary_loss_mlp": 0.01024036,
      "balance_loss_clip": 1.01222813,
      "balance_loss_mlp": 1.01691949,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 1.966950688182901,
      "language_loss": 0.77583337,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79781574,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.8413171768188477
    },
    {
      "auxiliary_loss_clip": 0.01162675,
      "auxiliary_loss_mlp": 0.01026397,
      "balance_loss_clip": 0.89276063,
      "balance_loss_mlp": 1.01875246,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.6548630633492856,
      "language_loss": 0.75443077,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77632147,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.7431836128234863
    },
    {
      "auxiliary_loss_clip": 0.01168994,
      "auxiliary_loss_mlp": 0.01022935,
      "balance_loss_clip": 1.01003027,
      "balance_loss_mlp": 1.01644969,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 2.0079595301198525,
      "language_loss": 0.78433609,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.80625534,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 2.5968751907348633
    },
    {
      "auxiliary_loss_clip": 0.01166694,
      "auxiliary_loss_mlp": 0.01026032,
      "balance_loss_clip": 1.01025355,
      "balance_loss_mlp": 1.01910567,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 1.6230270305987595,
      "language_loss": 0.88327885,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90520608,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.677717685699463
    },
    {
      "auxiliary_loss_clip": 0.01166128,
      "auxiliary_loss_mlp": 0.01030533,
      "balance_loss_clip": 1.0112803,
      "balance_loss_mlp": 1.02350855,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 2.004652453659147,
      "language_loss": 0.69583356,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.71780014,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 2.6670029163360596
    },
    {
      "auxiliary_loss_clip": 0.01170025,
      "auxiliary_loss_mlp": 0.01025615,
      "balance_loss_clip": 1.00981724,
      "balance_loss_mlp": 1.01791453,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 2.011961271409599,
      "language_loss": 0.74690199,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.76885843,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.7872517108917236
    },
    {
      "auxiliary_loss_clip": 0.01169233,
      "auxiliary_loss_mlp": 0.01025013,
      "balance_loss_clip": 1.00979769,
      "balance_loss_mlp": 1.01839685,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 2.2742203305600834,
      "language_loss": 0.79823643,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.82017887,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 2.6525237560272217
    },
    {
      "auxiliary_loss_clip": 0.01163252,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 0.93246746,
      "balance_loss_mlp": 1.01781154,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 2.2944807786320593,
      "language_loss": 0.78569186,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.80757761,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.7829432487487793
    },
    {
      "auxiliary_loss_clip": 0.01170724,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 0.81971085,
      "balance_loss_mlp": 1.0175966,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 1.7350096084025706,
      "language_loss": 0.70779735,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.72975659,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 2.9193215370178223
    },
    {
      "auxiliary_loss_clip": 0.01165731,
      "auxiliary_loss_mlp": 0.01122703,
      "balance_loss_clip": 1.01077092,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 1.6904891265534132,
      "language_loss": 0.73540616,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75829053,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 3.0810165405273438
    },
    {
      "auxiliary_loss_clip": 0.01168623,
      "auxiliary_loss_mlp": 0.0102431,
      "balance_loss_clip": 1.04789305,
      "balance_loss_mlp": 1.01708615,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 1.596986932201288,
      "language_loss": 0.7040844,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72601378,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 2.7160420417785645
    },
    {
      "auxiliary_loss_clip": 0.01172794,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.01195431,
      "balance_loss_mlp": 1.02066147,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 2.7540142384907917,
      "language_loss": 0.65140307,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67340636,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.7826123237609863
    },
    {
      "auxiliary_loss_clip": 0.01169497,
      "auxiliary_loss_mlp": 0.01026714,
      "balance_loss_clip": 0.97360063,
      "balance_loss_mlp": 1.01935911,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 1.8527087445229178,
      "language_loss": 0.74258101,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76454312,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.6931209564208984
    },
    {
      "auxiliary_loss_clip": 0.01166698,
      "auxiliary_loss_mlp": 0.01122757,
      "balance_loss_clip": 0.97528207,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 1.920579024467232,
      "language_loss": 0.73254132,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.75543588,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 2.6096713542938232
    },
    {
      "auxiliary_loss_clip": 0.01166636,
      "auxiliary_loss_mlp": 0.0103061,
      "balance_loss_clip": 0.97278845,
      "balance_loss_mlp": 1.0236783,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.8295906120720167,
      "language_loss": 0.80272591,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82469839,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.6913812160491943
    },
    {
      "auxiliary_loss_clip": 0.01172215,
      "auxiliary_loss_mlp": 0.01026886,
      "balance_loss_clip": 0.97179788,
      "balance_loss_mlp": 1.01998067,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.5390560132150037,
      "language_loss": 0.77242267,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79441369,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 4.009011268615723
    },
    {
      "auxiliary_loss_clip": 0.01164263,
      "auxiliary_loss_mlp": 0.01030264,
      "balance_loss_clip": 0.82402188,
      "balance_loss_mlp": 1.02269983,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 3.582391477923609,
      "language_loss": 0.65569842,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.67764366,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.824582576751709
    },
    {
      "auxiliary_loss_clip": 0.01174396,
      "auxiliary_loss_mlp": 0.01023856,
      "balance_loss_clip": 0.97348696,
      "balance_loss_mlp": 1.01693892,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 3.8162064878892017,
      "language_loss": 0.67093498,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69291747,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.689446449279785
    },
    {
      "auxiliary_loss_clip": 0.01171846,
      "auxiliary_loss_mlp": 0.01030111,
      "balance_loss_clip": 0.93552297,
      "balance_loss_mlp": 1.02244616,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 1.904122001832646,
      "language_loss": 0.78787041,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.80989003,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.758591413497925
    },
    {
      "auxiliary_loss_clip": 0.01174834,
      "auxiliary_loss_mlp": 0.010247,
      "balance_loss_clip": 0.93657029,
      "balance_loss_mlp": 1.01766682,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 2.250993433735692,
      "language_loss": 0.67705584,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.69905114,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.711188316345215
    },
    {
      "auxiliary_loss_clip": 0.01070852,
      "auxiliary_loss_mlp": 0.01115982,
      "balance_loss_clip": 0.86186337,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7529036303924393,
      "language_loss": 0.55826175,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.5801301,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.288889169692993
    },
    {
      "auxiliary_loss_clip": 0.01169313,
      "auxiliary_loss_mlp": 0.01021675,
      "balance_loss_clip": 0.97314227,
      "balance_loss_mlp": 1.01432872,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.878371774100965,
      "language_loss": 0.76999062,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.7919004,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 4.431893825531006
    },
    {
      "auxiliary_loss_clip": 0.01170442,
      "auxiliary_loss_mlp": 0.01022849,
      "balance_loss_clip": 1.05010641,
      "balance_loss_mlp": 1.01514196,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 2.5955382611898603,
      "language_loss": 0.774822,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.79675484,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 3.569495916366577
    },
    {
      "auxiliary_loss_clip": 0.01166145,
      "auxiliary_loss_mlp": 0.01031405,
      "balance_loss_clip": 0.97426558,
      "balance_loss_mlp": 1.02393603,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.026987188461399,
      "language_loss": 0.82505465,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.8470301,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 2.613504648208618
    },
    {
      "auxiliary_loss_clip": 0.01145857,
      "auxiliary_loss_mlp": 0.01026186,
      "balance_loss_clip": 0.81501549,
      "balance_loss_mlp": 1.01861036,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 1.8590624164288236,
      "language_loss": 0.74720961,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.76893002,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.8600378036499023
    },
    {
      "auxiliary_loss_clip": 0.01152838,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 0.85076618,
      "balance_loss_mlp": 1.02077198,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 4.07319428323177,
      "language_loss": 0.78068614,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80249435,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 2.8443751335144043
    },
    {
      "auxiliary_loss_clip": 0.01172676,
      "auxiliary_loss_mlp": 0.01024602,
      "balance_loss_clip": 0.93361658,
      "balance_loss_mlp": 1.01612616,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.1989188589040816,
      "language_loss": 0.71867955,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.74065232,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.743403673171997
    },
    {
      "auxiliary_loss_clip": 0.01168392,
      "auxiliary_loss_mlp": 0.01027705,
      "balance_loss_clip": 1.01277804,
      "balance_loss_mlp": 1.02085936,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 1.9847235348334233,
      "language_loss": 0.73896283,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.7609238,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.7661936283111572
    },
    {
      "auxiliary_loss_clip": 0.01162106,
      "auxiliary_loss_mlp": 0.01029826,
      "balance_loss_clip": 0.97019291,
      "balance_loss_mlp": 1.0223577,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 2.035974448678337,
      "language_loss": 0.62690818,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.64882743,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.7271385192871094
    },
    {
      "auxiliary_loss_clip": 0.01160641,
      "auxiliary_loss_mlp": 0.01122493,
      "balance_loss_clip": 0.89377481,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 2.186908743616244,
      "language_loss": 0.753802,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.77663338,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 2.7433953285217285
    },
    {
      "auxiliary_loss_clip": 0.011703,
      "auxiliary_loss_mlp": 0.01025363,
      "balance_loss_clip": 0.97044599,
      "balance_loss_mlp": 1.01772749,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.56203817178192,
      "language_loss": 0.78287977,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80483639,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.6492655277252197
    },
    {
      "auxiliary_loss_clip": 0.01162969,
      "auxiliary_loss_mlp": 0.01028893,
      "balance_loss_clip": 1.00923002,
      "balance_loss_mlp": 1.02116835,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 1.7861839619353226,
      "language_loss": 0.71355957,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.73547816,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 2.6849591732025146
    },
    {
      "auxiliary_loss_clip": 0.01167173,
      "auxiliary_loss_mlp": 0.01029276,
      "balance_loss_clip": 0.93477273,
      "balance_loss_mlp": 1.02184296,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.013261649342481,
      "language_loss": 0.62338996,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.64535445,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.7785449028015137
    },
    {
      "auxiliary_loss_clip": 0.01172407,
      "auxiliary_loss_mlp": 0.0103142,
      "balance_loss_clip": 0.97345626,
      "balance_loss_mlp": 1.02399623,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 2.2577887751550603,
      "language_loss": 0.79424095,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.81627923,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 2.7091352939605713
    },
    {
      "auxiliary_loss_clip": 0.01173478,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 0.93602359,
      "balance_loss_mlp": 1.02492571,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 1.715683474359595,
      "language_loss": 0.71597636,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.73803174,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.750312089920044
    },
    {
      "auxiliary_loss_clip": 0.01064988,
      "auxiliary_loss_mlp": 0.01002122,
      "balance_loss_clip": 1.0112021,
      "balance_loss_mlp": 1.00048876,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.7941691167042878,
      "language_loss": 0.62375724,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64442831,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 3.25520658493042
    },
    {
      "auxiliary_loss_clip": 0.01173563,
      "auxiliary_loss_mlp": 0.01026506,
      "balance_loss_clip": 1.01504183,
      "balance_loss_mlp": 1.0192219,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 1.8386399311957164,
      "language_loss": 0.66937912,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.69137979,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 2.649766683578491
    },
    {
      "auxiliary_loss_clip": 0.01176445,
      "auxiliary_loss_mlp": 0.01027774,
      "balance_loss_clip": 1.04966629,
      "balance_loss_mlp": 1.0196023,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 1.9841884551136904,
      "language_loss": 0.79673028,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.81877244,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.6607861518859863
    },
    {
      "auxiliary_loss_clip": 0.01172392,
      "auxiliary_loss_mlp": 0.01022048,
      "balance_loss_clip": 1.0116719,
      "balance_loss_mlp": 1.01484144,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 1.9998268636434067,
      "language_loss": 0.72944009,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.7513845,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.6604793071746826
    },
    {
      "auxiliary_loss_clip": 0.01169219,
      "auxiliary_loss_mlp": 0.01032182,
      "balance_loss_clip": 0.89397204,
      "balance_loss_mlp": 1.02449632,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 1.9580734528840984,
      "language_loss": 0.72569036,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74770439,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.711390733718872
    },
    {
      "auxiliary_loss_clip": 0.01169561,
      "auxiliary_loss_mlp": 0.01027405,
      "balance_loss_clip": 0.85679674,
      "balance_loss_mlp": 1.02060974,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.4408927493891752,
      "language_loss": 0.76030707,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78227675,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 2.8167552947998047
    },
    {
      "auxiliary_loss_clip": 0.01161622,
      "auxiliary_loss_mlp": 0.01028525,
      "balance_loss_clip": 0.89243841,
      "balance_loss_mlp": 1.0208714,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 1.8739460583281569,
      "language_loss": 0.82058489,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.84248632,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 3.78566575050354
    },
    {
      "auxiliary_loss_clip": 0.0117096,
      "auxiliary_loss_mlp": 0.0102418,
      "balance_loss_clip": 1.04803765,
      "balance_loss_mlp": 1.0163238,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 2.42301461340185,
      "language_loss": 0.8423388,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.86429018,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.6560187339782715
    },
    {
      "auxiliary_loss_clip": 0.01157217,
      "auxiliary_loss_mlp": 0.010302,
      "balance_loss_clip": 0.92988414,
      "balance_loss_mlp": 1.02246881,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 1.9742853453634266,
      "language_loss": 0.75579369,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77766788,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.7428526878356934
    },
    {
      "auxiliary_loss_clip": 0.01157466,
      "auxiliary_loss_mlp": 0.01023163,
      "balance_loss_clip": 0.93199211,
      "balance_loss_mlp": 1.01563215,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.0817218197950487,
      "language_loss": 0.73989373,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.76170003,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.753983497619629
    },
    {
      "auxiliary_loss_clip": 0.01172287,
      "auxiliary_loss_mlp": 0.0102336,
      "balance_loss_clip": 1.05079722,
      "balance_loss_mlp": 1.01648188,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.552279912227689,
      "language_loss": 0.8270396,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.8489961,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.637824296951294
    },
    {
      "auxiliary_loss_clip": 0.0116909,
      "auxiliary_loss_mlp": 0.01027505,
      "balance_loss_clip": 1.00824475,
      "balance_loss_mlp": 1.02016747,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.231658529873091,
      "language_loss": 0.77496445,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.79693043,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 3.51859450340271
    },
    {
      "auxiliary_loss_clip": 0.01161899,
      "auxiliary_loss_mlp": 0.0102102,
      "balance_loss_clip": 0.97239602,
      "balance_loss_mlp": 1.01382565,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 1.7092363144492346,
      "language_loss": 0.7635746,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.78540379,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 3.747009515762329
    },
    {
      "auxiliary_loss_clip": 0.0116625,
      "auxiliary_loss_mlp": 0.01024963,
      "balance_loss_clip": 1.01127815,
      "balance_loss_mlp": 1.01805472,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 7.6350717216944615,
      "language_loss": 0.80390704,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.82581925,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 3.6035659313201904
    },
    {
      "auxiliary_loss_clip": 0.01169576,
      "auxiliary_loss_mlp": 0.01031601,
      "balance_loss_clip": 1.01154661,
      "balance_loss_mlp": 1.02434683,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 1.6301122502483503,
      "language_loss": 0.64513677,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66714853,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.8734097480773926
    },
    {
      "auxiliary_loss_clip": 0.01168539,
      "auxiliary_loss_mlp": 0.0102498,
      "balance_loss_clip": 0.97160566,
      "balance_loss_mlp": 1.01808333,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 2.0839169569069673,
      "language_loss": 0.69272655,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.71466178,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.6640939712524414
    },
    {
      "auxiliary_loss_clip": 0.01165953,
      "auxiliary_loss_mlp": 0.01123117,
      "balance_loss_clip": 0.78170496,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 4.607776806713373,
      "language_loss": 0.78445661,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.8073473,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 2.9818103313446045
    },
    {
      "auxiliary_loss_clip": 0.01163121,
      "auxiliary_loss_mlp": 0.01029328,
      "balance_loss_clip": 0.93465424,
      "balance_loss_mlp": 1.02171659,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.149886572731091,
      "language_loss": 0.81241554,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83434004,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 3.4884440898895264
    },
    {
      "auxiliary_loss_clip": 0.01172338,
      "auxiliary_loss_mlp": 0.01029729,
      "balance_loss_clip": 1.01121402,
      "balance_loss_mlp": 1.02248979,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 1.7867383118866773,
      "language_loss": 0.74513018,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.76715088,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.79034161567688
    },
    {
      "auxiliary_loss_clip": 0.01164513,
      "auxiliary_loss_mlp": 0.01022684,
      "balance_loss_clip": 0.97069633,
      "balance_loss_mlp": 1.015517,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 1.9556092053341714,
      "language_loss": 0.69520903,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71708101,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.7453815937042236
    },
    {
      "auxiliary_loss_clip": 0.011538,
      "auxiliary_loss_mlp": 0.01026594,
      "balance_loss_clip": 0.92968225,
      "balance_loss_mlp": 1.01913738,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 1.6185271372067218,
      "language_loss": 0.72078145,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74258536,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 2.716728448867798
    },
    {
      "auxiliary_loss_clip": 0.01153871,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 0.89615571,
      "balance_loss_mlp": 1.02108884,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 1.783948804590144,
      "language_loss": 0.78969872,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.81152338,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.8618764877319336
    },
    {
      "auxiliary_loss_clip": 0.01168683,
      "auxiliary_loss_mlp": 0.01023357,
      "balance_loss_clip": 1.04977918,
      "balance_loss_mlp": 1.01616311,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.8340813721520082,
      "language_loss": 0.7446245,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.76654482,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 2.6497247219085693
    },
    {
      "auxiliary_loss_clip": 0.01170315,
      "auxiliary_loss_mlp": 0.01122103,
      "balance_loss_clip": 0.89817512,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.6646444159863294,
      "language_loss": 0.80386019,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.82678431,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 2.7790586948394775
    },
    {
      "auxiliary_loss_clip": 0.01172238,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 0.97141761,
      "balance_loss_mlp": 1.01932657,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 1.6015180122888437,
      "language_loss": 0.75703299,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.77901894,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 2.701676845550537
    },
    {
      "auxiliary_loss_clip": 0.01169138,
      "auxiliary_loss_mlp": 0.01028346,
      "balance_loss_clip": 0.9752143,
      "balance_loss_mlp": 1.02148223,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.770286259325702,
      "language_loss": 0.71724266,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.73921752,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.685312032699585
    },
    {
      "auxiliary_loss_clip": 0.01172754,
      "auxiliary_loss_mlp": 0.01026494,
      "balance_loss_clip": 0.97183514,
      "balance_loss_mlp": 1.01866174,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 1.6209848438253913,
      "language_loss": 0.68994021,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71193278,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 2.860853433609009
    },
    {
      "auxiliary_loss_clip": 0.01159585,
      "auxiliary_loss_mlp": 0.01027759,
      "balance_loss_clip": 1.00837493,
      "balance_loss_mlp": 1.02004623,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 1.7856862696730422,
      "language_loss": 0.79621798,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.81809139,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 2.6197242736816406
    },
    {
      "auxiliary_loss_clip": 0.01168404,
      "auxiliary_loss_mlp": 0.01026203,
      "balance_loss_clip": 1.00921321,
      "balance_loss_mlp": 1.01850224,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 1.9762641618081038,
      "language_loss": 0.77778184,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.79972792,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.6298670768737793
    },
    {
      "auxiliary_loss_clip": 0.0106985,
      "auxiliary_loss_mlp": 0.01003715,
      "balance_loss_clip": 0.86422467,
      "balance_loss_mlp": 1.00192726,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7763273066117955,
      "language_loss": 0.64278716,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.66352284,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.399152994155884
    },
    {
      "auxiliary_loss_clip": 0.01163356,
      "auxiliary_loss_mlp": 0.01022654,
      "balance_loss_clip": 0.97231877,
      "balance_loss_mlp": 1.01604056,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 3.4110806643419367,
      "language_loss": 0.73340219,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.75526226,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.6342203617095947
    },
    {
      "auxiliary_loss_clip": 0.01169229,
      "auxiliary_loss_mlp": 0.01022481,
      "balance_loss_clip": 0.97202861,
      "balance_loss_mlp": 1.01512313,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 1.9262015584175645,
      "language_loss": 0.63676572,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65868282,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.726402521133423
    },
    {
      "auxiliary_loss_clip": 0.01171655,
      "auxiliary_loss_mlp": 0.01021376,
      "balance_loss_clip": 0.97086471,
      "balance_loss_mlp": 1.01418781,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.033279171575098,
      "language_loss": 0.72460526,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.74653554,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 3.734870433807373
    },
    {
      "auxiliary_loss_clip": 0.01161891,
      "auxiliary_loss_mlp": 0.01025104,
      "balance_loss_clip": 0.89475054,
      "balance_loss_mlp": 1.01755786,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 1.981066104911819,
      "language_loss": 0.74687165,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.76874161,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.6650161743164062
    },
    {
      "auxiliary_loss_clip": 0.01169487,
      "auxiliary_loss_mlp": 0.0102716,
      "balance_loss_clip": 1.04838073,
      "balance_loss_mlp": 1.02020717,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.1161386779564304,
      "language_loss": 0.61328065,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63524711,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.6187477111816406
    },
    {
      "auxiliary_loss_clip": 0.01170113,
      "auxiliary_loss_mlp": 0.0103342,
      "balance_loss_clip": 0.93704432,
      "balance_loss_mlp": 1.02579987,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 1.8804740309071766,
      "language_loss": 0.71334946,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.73538476,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.752934455871582
    },
    {
      "auxiliary_loss_clip": 0.01162938,
      "auxiliary_loss_mlp": 0.01026204,
      "balance_loss_clip": 0.97238445,
      "balance_loss_mlp": 1.0190959,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.646135833218756,
      "language_loss": 0.71740401,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.73929542,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.81315279006958
    },
    {
      "auxiliary_loss_clip": 0.01170215,
      "auxiliary_loss_mlp": 0.01027415,
      "balance_loss_clip": 0.97142017,
      "balance_loss_mlp": 1.02028346,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.701229674930598,
      "language_loss": 0.80382133,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82579768,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.677824020385742
    },
    {
      "auxiliary_loss_clip": 0.01171311,
      "auxiliary_loss_mlp": 0.01023614,
      "balance_loss_clip": 0.93562198,
      "balance_loss_mlp": 1.01671791,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 1.9901152542192784,
      "language_loss": 0.81818771,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.84013689,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 3.6109368801116943
    },
    {
      "auxiliary_loss_clip": 0.0116477,
      "auxiliary_loss_mlp": 0.01024158,
      "balance_loss_clip": 0.97208792,
      "balance_loss_mlp": 1.01688051,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 3.335231407741031,
      "language_loss": 0.67571008,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.69759929,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 3.5523483753204346
    },
    {
      "auxiliary_loss_clip": 0.01158599,
      "auxiliary_loss_mlp": 0.01026004,
      "balance_loss_clip": 1.00642347,
      "balance_loss_mlp": 1.01864839,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 1.9473945711620657,
      "language_loss": 0.79466164,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.8165077,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 2.657966136932373
    },
    {
      "auxiliary_loss_clip": 0.01169279,
      "auxiliary_loss_mlp": 0.01024135,
      "balance_loss_clip": 1.01258945,
      "balance_loss_mlp": 1.01780462,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 2.036906537045561,
      "language_loss": 0.74693274,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.76886678,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.610193967819214
    },
    {
      "auxiliary_loss_clip": 0.0106906,
      "auxiliary_loss_mlp": 0.01001937,
      "balance_loss_clip": 0.93648714,
      "balance_loss_mlp": 1.0003041,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.8892427107516558,
      "language_loss": 0.60984886,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.63055885,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 3.1546268463134766
    },
    {
      "auxiliary_loss_clip": 0.01172768,
      "auxiliary_loss_mlp": 0.01026519,
      "balance_loss_clip": 1.01388061,
      "balance_loss_mlp": 1.01881766,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 2.2646465038586605,
      "language_loss": 0.83534312,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.85733604,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 2.5655906200408936
    },
    {
      "auxiliary_loss_clip": 0.01172894,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 0.89981669,
      "balance_loss_mlp": 1.02057648,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.46528575213011,
      "language_loss": 0.77539623,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.79740703,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.7123289108276367
    },
    {
      "auxiliary_loss_clip": 0.01167537,
      "auxiliary_loss_mlp": 0.01027321,
      "balance_loss_clip": 1.04638195,
      "balance_loss_mlp": 1.02013886,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 1.890333123041085,
      "language_loss": 0.73359251,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75554115,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.766113519668579
    },
    {
      "auxiliary_loss_clip": 0.01163206,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 0.93290311,
      "balance_loss_mlp": 1.01850319,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 3.090200910966046,
      "language_loss": 0.71581727,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73771071,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 2.701481819152832
    },
    {
      "auxiliary_loss_clip": 0.01164168,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 0.86091858,
      "balance_loss_mlp": 1.01722062,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 1.925268549922729,
      "language_loss": 0.82768464,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.84956968,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.7278621196746826
    },
    {
      "auxiliary_loss_clip": 0.01164693,
      "auxiliary_loss_mlp": 0.01122798,
      "balance_loss_clip": 1.00862098,
      "balance_loss_mlp": 0.0,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 2.12747691017428,
      "language_loss": 0.66559947,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68847442,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 2.6452839374542236
    },
    {
      "auxiliary_loss_clip": 0.01171923,
      "auxiliary_loss_mlp": 0.01027946,
      "balance_loss_clip": 0.93374604,
      "balance_loss_mlp": 1.02021551,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.7534520254675976,
      "language_loss": 0.75541687,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.77741551,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 2.7362258434295654
    },
    {
      "auxiliary_loss_clip": 0.01164323,
      "auxiliary_loss_mlp": 0.01024756,
      "balance_loss_clip": 0.8574183,
      "balance_loss_mlp": 1.01718044,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 1.9319822361127235,
      "language_loss": 0.71868598,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.74057674,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.7819225788116455
    },
    {
      "auxiliary_loss_clip": 0.01171943,
      "auxiliary_loss_mlp": 0.0112228,
      "balance_loss_clip": 0.97498691,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.761421148854184,
      "language_loss": 0.73617792,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.75912011,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 2.6875076293945312
    },
    {
      "auxiliary_loss_clip": 0.01167317,
      "auxiliary_loss_mlp": 0.01121702,
      "balance_loss_clip": 0.97114223,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 1.7386191609173802,
      "language_loss": 0.86436558,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.88725573,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 2.7231693267822266
    },
    {
      "auxiliary_loss_clip": 0.01169503,
      "auxiliary_loss_mlp": 0.01032844,
      "balance_loss_clip": 1.0494523,
      "balance_loss_mlp": 1.02560186,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 2.2375422019648363,
      "language_loss": 0.80207342,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82409692,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 2.655046224594116
    },
    {
      "auxiliary_loss_clip": 0.01167597,
      "auxiliary_loss_mlp": 0.01029322,
      "balance_loss_clip": 0.93200755,
      "balance_loss_mlp": 1.02249479,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 1.6108298463616646,
      "language_loss": 0.77431166,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.79628086,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.690657377243042
    },
    {
      "auxiliary_loss_clip": 0.01160108,
      "auxiliary_loss_mlp": 0.0101962,
      "balance_loss_clip": 0.96938497,
      "balance_loss_mlp": 1.01229215,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 2.328794375436072,
      "language_loss": 0.66859907,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69039631,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.72124981880188
    },
    {
      "auxiliary_loss_clip": 0.01160618,
      "auxiliary_loss_mlp": 0.01023914,
      "balance_loss_clip": 0.97307074,
      "balance_loss_mlp": 1.01664186,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 2.079206698687657,
      "language_loss": 0.7641291,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78597438,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.684319496154785
    },
    {
      "auxiliary_loss_clip": 0.01165788,
      "auxiliary_loss_mlp": 0.01022622,
      "balance_loss_clip": 0.89320838,
      "balance_loss_mlp": 1.01559794,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 6.114959794603954,
      "language_loss": 0.78827691,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.81016105,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.796278953552246
    },
    {
      "auxiliary_loss_clip": 0.01166239,
      "auxiliary_loss_mlp": 0.01024794,
      "balance_loss_clip": 0.93358994,
      "balance_loss_mlp": 1.01747417,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 1.9717724440883153,
      "language_loss": 0.70304692,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72495723,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 3.7383270263671875
    },
    {
      "auxiliary_loss_clip": 0.01158753,
      "auxiliary_loss_mlp": 0.0102457,
      "balance_loss_clip": 0.9702211,
      "balance_loss_mlp": 1.01765013,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 12.005205583506507,
      "language_loss": 0.78907949,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.81091273,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.668665647506714
    },
    {
      "auxiliary_loss_clip": 0.01170982,
      "auxiliary_loss_mlp": 0.01122566,
      "balance_loss_clip": 0.97444516,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 1.8704651115664788,
      "language_loss": 0.8578155,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.88075101,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.7378733158111572
    },
    {
      "auxiliary_loss_clip": 0.01170832,
      "auxiliary_loss_mlp": 0.01023226,
      "balance_loss_clip": 1.01185656,
      "balance_loss_mlp": 1.01565671,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 26.21147181818955,
      "language_loss": 0.74025989,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76220047,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.625521421432495
    },
    {
      "auxiliary_loss_clip": 0.01172513,
      "auxiliary_loss_mlp": 0.01024304,
      "balance_loss_clip": 0.97480989,
      "balance_loss_mlp": 1.01665354,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 2.2432732975157976,
      "language_loss": 0.76657844,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.78854656,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.677467107772827
    },
    {
      "auxiliary_loss_clip": 0.0116893,
      "auxiliary_loss_mlp": 0.01028547,
      "balance_loss_clip": 1.04943967,
      "balance_loss_mlp": 1.02135551,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 1.5633382636058244,
      "language_loss": 0.72672129,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.74869609,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 3.4619925022125244
    },
    {
      "auxiliary_loss_clip": 0.01168691,
      "auxiliary_loss_mlp": 0.01023243,
      "balance_loss_clip": 1.04875338,
      "balance_loss_mlp": 1.01619148,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 1.7870486009265363,
      "language_loss": 0.73804379,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.7599631,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 3.5681838989257812
    },
    {
      "auxiliary_loss_clip": 0.01173893,
      "auxiliary_loss_mlp": 0.01021776,
      "balance_loss_clip": 0.97550303,
      "balance_loss_mlp": 1.01409924,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 2.0277274732511215,
      "language_loss": 0.84717095,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.86912763,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.6742212772369385
    },
    {
      "auxiliary_loss_clip": 0.01169144,
      "auxiliary_loss_mlp": 0.01027585,
      "balance_loss_clip": 1.00976682,
      "balance_loss_mlp": 1.02023327,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 1.6293353318030808,
      "language_loss": 0.86733997,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.88930726,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 3.7582955360412598
    },
    {
      "auxiliary_loss_clip": 0.0116828,
      "auxiliary_loss_mlp": 0.01027139,
      "balance_loss_clip": 0.97071612,
      "balance_loss_mlp": 1.0196588,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 1.765440173020377,
      "language_loss": 0.60055315,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62250733,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.643012523651123
    },
    {
      "auxiliary_loss_clip": 0.01164814,
      "auxiliary_loss_mlp": 0.0103247,
      "balance_loss_clip": 0.97114694,
      "balance_loss_mlp": 1.02577937,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 1.8427055331437334,
      "language_loss": 0.80780947,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.82978237,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.683169364929199
    },
    {
      "auxiliary_loss_clip": 0.01171115,
      "auxiliary_loss_mlp": 0.01028829,
      "balance_loss_clip": 0.97569716,
      "balance_loss_mlp": 1.02167034,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 1.8034823091787358,
      "language_loss": 0.86817098,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.8901704,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.697247266769409
    },
    {
      "auxiliary_loss_clip": 0.01173425,
      "auxiliary_loss_mlp": 0.01023309,
      "balance_loss_clip": 0.93528545,
      "balance_loss_mlp": 1.01597142,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 1.9149572688514118,
      "language_loss": 0.75431573,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77628309,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 3.005363702774048
    },
    {
      "auxiliary_loss_clip": 0.01068765,
      "auxiliary_loss_mlp": 0.01002346,
      "balance_loss_clip": 0.97404838,
      "balance_loss_mlp": 1.00079656,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.8936855233646073,
      "language_loss": 0.60206068,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.62277174,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 3.0834691524505615
    },
    {
      "auxiliary_loss_clip": 0.01161852,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 0.97013462,
      "balance_loss_mlp": 1.022156,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 5.736248161745113,
      "language_loss": 0.68939614,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.71131134,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 2.747832775115967
    },
    {
      "auxiliary_loss_clip": 0.01172655,
      "auxiliary_loss_mlp": 0.01024903,
      "balance_loss_clip": 1.04982018,
      "balance_loss_mlp": 1.01721954,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 1.863545371536794,
      "language_loss": 0.75057042,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.77254593,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.585859537124634
    },
    {
      "auxiliary_loss_clip": 0.01171505,
      "auxiliary_loss_mlp": 0.01023792,
      "balance_loss_clip": 1.00936556,
      "balance_loss_mlp": 1.01662505,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 1.8027704031056198,
      "language_loss": 0.72109282,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.74304581,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.71332049369812
    },
    {
      "auxiliary_loss_clip": 0.01071827,
      "auxiliary_loss_mlp": 0.01000683,
      "balance_loss_clip": 0.86212826,
      "balance_loss_mlp": 0.99905014,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.8722156875458177,
      "language_loss": 0.5387423,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55946738,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 3.22176194190979
    },
    {
      "auxiliary_loss_clip": 0.01166081,
      "auxiliary_loss_mlp": 0.01023732,
      "balance_loss_clip": 1.0475167,
      "balance_loss_mlp": 1.01666248,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 1.7902930048958947,
      "language_loss": 0.81092465,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.83282274,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 2.9334218502044678
    },
    {
      "auxiliary_loss_clip": 0.01170734,
      "auxiliary_loss_mlp": 0.0102597,
      "balance_loss_clip": 1.01264882,
      "balance_loss_mlp": 1.01853752,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 1.5288632838361256,
      "language_loss": 0.77975583,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.80172288,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.6849265098571777
    },
    {
      "auxiliary_loss_clip": 0.01159264,
      "auxiliary_loss_mlp": 0.01024637,
      "balance_loss_clip": 0.89495695,
      "balance_loss_mlp": 1.01728487,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 1.9749492682988228,
      "language_loss": 0.74300039,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76483941,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 2.7145891189575195
    },
    {
      "auxiliary_loss_clip": 0.01146129,
      "auxiliary_loss_mlp": 0.0102185,
      "balance_loss_clip": 0.85401022,
      "balance_loss_mlp": 1.01496005,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 1.8595927569596427,
      "language_loss": 0.6777401,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69941998,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 2.8325469493865967
    },
    {
      "auxiliary_loss_clip": 0.01170573,
      "auxiliary_loss_mlp": 0.0112243,
      "balance_loss_clip": 1.05184472,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 1.5793574244879,
      "language_loss": 0.74397683,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.76690686,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.6157374382019043
    },
    {
      "auxiliary_loss_clip": 0.01165101,
      "auxiliary_loss_mlp": 0.01028902,
      "balance_loss_clip": 0.90043849,
      "balance_loss_mlp": 1.02093923,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 2.22656331979498,
      "language_loss": 0.83387029,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85581034,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.7490711212158203
    },
    {
      "auxiliary_loss_clip": 0.01159853,
      "auxiliary_loss_mlp": 0.01028176,
      "balance_loss_clip": 1.00816798,
      "balance_loss_mlp": 1.02068412,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 2.24392780102091,
      "language_loss": 0.80395216,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82583249,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.694457530975342
    },
    {
      "auxiliary_loss_clip": 0.01169302,
      "auxiliary_loss_mlp": 0.01025922,
      "balance_loss_clip": 1.00917244,
      "balance_loss_mlp": 1.01892447,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 2.005451332282776,
      "language_loss": 0.79433179,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.816284,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.676502227783203
    },
    {
      "auxiliary_loss_clip": 0.01167221,
      "auxiliary_loss_mlp": 0.01122364,
      "balance_loss_clip": 1.00970268,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 2.0501751050764097,
      "language_loss": 0.63876891,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.66166472,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.8017098903656006
    },
    {
      "auxiliary_loss_clip": 0.01165165,
      "auxiliary_loss_mlp": 0.01024916,
      "balance_loss_clip": 0.93454814,
      "balance_loss_mlp": 1.01723349,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 2.0113066319937722,
      "language_loss": 0.81009305,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.83199382,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 3.695596218109131
    },
    {
      "auxiliary_loss_clip": 0.0116127,
      "auxiliary_loss_mlp": 0.0102251,
      "balance_loss_clip": 0.96843386,
      "balance_loss_mlp": 1.01502991,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.804858232154143,
      "language_loss": 0.761724,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78356177,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 2.7079105377197266
    },
    {
      "auxiliary_loss_clip": 0.01169318,
      "auxiliary_loss_mlp": 0.0102607,
      "balance_loss_clip": 1.0508213,
      "balance_loss_mlp": 1.01930451,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 1.8859810350697823,
      "language_loss": 0.72783035,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.74978417,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.606036424636841
    },
    {
      "auxiliary_loss_clip": 0.01161952,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 0.93129969,
      "balance_loss_mlp": 1.01784134,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 1.814467482135841,
      "language_loss": 0.70588779,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.72775942,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.7422268390655518
    },
    {
      "auxiliary_loss_clip": 0.01163963,
      "auxiliary_loss_mlp": 0.01023466,
      "balance_loss_clip": 0.93556166,
      "balance_loss_mlp": 1.01634288,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.110151686011257,
      "language_loss": 0.85399419,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.87586844,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 3.5479886531829834
    },
    {
      "auxiliary_loss_clip": 0.01162039,
      "auxiliary_loss_mlp": 0.01026845,
      "balance_loss_clip": 0.97016191,
      "balance_loss_mlp": 1.01932633,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 1.733217632454185,
      "language_loss": 0.68206221,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.703951,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 3.67118501663208
    },
    {
      "auxiliary_loss_clip": 0.01164909,
      "auxiliary_loss_mlp": 0.01022834,
      "balance_loss_clip": 0.96739846,
      "balance_loss_mlp": 1.01558018,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 1.8048116703514037,
      "language_loss": 0.69536269,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.71724021,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.700701951980591
    },
    {
      "auxiliary_loss_clip": 0.01167151,
      "auxiliary_loss_mlp": 0.01020048,
      "balance_loss_clip": 1.00865102,
      "balance_loss_mlp": 1.0128572,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 2.0009251053413397,
      "language_loss": 0.76992285,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.79179484,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 3.588959217071533
    },
    {
      "auxiliary_loss_clip": 0.01172111,
      "auxiliary_loss_mlp": 0.01024176,
      "balance_loss_clip": 1.05124831,
      "balance_loss_mlp": 1.01661217,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.4627062214653765,
      "language_loss": 0.75201952,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.77398235,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.699378490447998
    },
    {
      "auxiliary_loss_clip": 0.01169876,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.00855184,
      "balance_loss_mlp": 1.01905668,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 1.9513912575807437,
      "language_loss": 0.62624812,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.64821106,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.6811249256134033
    },
    {
      "auxiliary_loss_clip": 0.01164912,
      "auxiliary_loss_mlp": 0.0102889,
      "balance_loss_clip": 0.93412459,
      "balance_loss_mlp": 1.02113581,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.049727221985915,
      "language_loss": 0.82609189,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.84802991,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.688426971435547
    },
    {
      "auxiliary_loss_clip": 0.0116528,
      "auxiliary_loss_mlp": 0.01028093,
      "balance_loss_clip": 1.0097717,
      "balance_loss_mlp": 1.02064788,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 2.0377221403983263,
      "language_loss": 0.77630901,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79824275,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.654358148574829
    },
    {
      "auxiliary_loss_clip": 0.0117033,
      "auxiliary_loss_mlp": 0.01025856,
      "balance_loss_clip": 1.01187444,
      "balance_loss_mlp": 1.01845336,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 1.732093918846895,
      "language_loss": 0.8101939,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.8321557,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 2.6006462574005127
    },
    {
      "auxiliary_loss_clip": 0.01162735,
      "auxiliary_loss_mlp": 0.010241,
      "balance_loss_clip": 0.93183291,
      "balance_loss_mlp": 1.01708746,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 1.7519494432369214,
      "language_loss": 0.78907704,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81094539,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 2.763079881668091
    },
    {
      "auxiliary_loss_clip": 0.01064952,
      "auxiliary_loss_mlp": 0.01002247,
      "balance_loss_clip": 1.01174664,
      "balance_loss_mlp": 1.00062561,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7554197944433955,
      "language_loss": 0.58339846,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60407043,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.2642672061920166
    },
    {
      "auxiliary_loss_clip": 0.01167868,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.0472033,
      "balance_loss_mlp": 1.02042401,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.0751797869437483,
      "language_loss": 0.78922772,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81118399,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.5308218002319336
    },
    {
      "auxiliary_loss_clip": 0.01166397,
      "auxiliary_loss_mlp": 0.01023703,
      "balance_loss_clip": 0.93204826,
      "balance_loss_mlp": 1.01661587,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 2.916334987349192,
      "language_loss": 0.85487068,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87677169,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 2.652102470397949
    },
    {
      "auxiliary_loss_clip": 0.01160635,
      "auxiliary_loss_mlp": 0.01029762,
      "balance_loss_clip": 0.96733153,
      "balance_loss_mlp": 1.02218676,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.8341740923302696,
      "language_loss": 0.78577006,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.80767399,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 2.6979880332946777
    },
    {
      "auxiliary_loss_clip": 0.01165441,
      "auxiliary_loss_mlp": 0.01028874,
      "balance_loss_clip": 0.97133374,
      "balance_loss_mlp": 1.02178144,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 1.7900098885975293,
      "language_loss": 0.73300624,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.75494945,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 2.678314685821533
    },
    {
      "auxiliary_loss_clip": 0.01167035,
      "auxiliary_loss_mlp": 0.01028087,
      "balance_loss_clip": 1.00987482,
      "balance_loss_mlp": 1.02074993,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 2.1298943967505783,
      "language_loss": 0.7585398,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78049099,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 2.688575506210327
    },
    {
      "auxiliary_loss_clip": 0.01164742,
      "auxiliary_loss_mlp": 0.01029221,
      "balance_loss_clip": 0.93577504,
      "balance_loss_mlp": 1.02211618,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 2.013938665170355,
      "language_loss": 0.7567144,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.77865404,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 2.68019437789917
    },
    {
      "auxiliary_loss_clip": 0.01165838,
      "auxiliary_loss_mlp": 0.01029603,
      "balance_loss_clip": 0.97482693,
      "balance_loss_mlp": 1.02178836,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.901523947995716,
      "language_loss": 0.74131054,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.76326495,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.6990363597869873
    },
    {
      "auxiliary_loss_clip": 0.01068144,
      "auxiliary_loss_mlp": 0.01000631,
      "balance_loss_clip": 0.97374415,
      "balance_loss_mlp": 0.99895042,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7471715401008999,
      "language_loss": 0.59536934,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61605716,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.2300100326538086
    },
    {
      "auxiliary_loss_clip": 0.01161155,
      "auxiliary_loss_mlp": 0.01029799,
      "balance_loss_clip": 0.96812975,
      "balance_loss_mlp": 1.02242005,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 2.0320157216357018,
      "language_loss": 0.70947063,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73138022,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.6501693725585938
    },
    {
      "auxiliary_loss_clip": 0.01169512,
      "auxiliary_loss_mlp": 0.0102774,
      "balance_loss_clip": 1.05043125,
      "balance_loss_mlp": 1.0202837,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.545518236840743,
      "language_loss": 0.75956267,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78153527,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.647714376449585
    },
    {
      "auxiliary_loss_clip": 0.01169755,
      "auxiliary_loss_mlp": 0.01029432,
      "balance_loss_clip": 1.01381874,
      "balance_loss_mlp": 1.02174044,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 2.1506242850049837,
      "language_loss": 0.76122266,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.78321457,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.659026861190796
    },
    {
      "auxiliary_loss_clip": 0.0116683,
      "auxiliary_loss_mlp": 0.01024703,
      "balance_loss_clip": 1.01036382,
      "balance_loss_mlp": 1.01770568,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.2654741074960305,
      "language_loss": 0.75715971,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.77907515,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 3.6192262172698975
    },
    {
      "auxiliary_loss_clip": 0.01167774,
      "auxiliary_loss_mlp": 0.01022519,
      "balance_loss_clip": 0.97241354,
      "balance_loss_mlp": 1.01560521,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.8255600878364793,
      "language_loss": 0.66227281,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68417573,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 2.8254430294036865
    },
    {
      "auxiliary_loss_clip": 0.01158241,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 0.9690659,
      "balance_loss_mlp": 1.01898921,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.5973954250596196,
      "language_loss": 0.8471874,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.86902994,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.6871860027313232
    },
    {
      "auxiliary_loss_clip": 0.011693,
      "auxiliary_loss_mlp": 0.01035168,
      "balance_loss_clip": 0.97469413,
      "balance_loss_mlp": 1.02722859,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.2518391862965537,
      "language_loss": 0.86422455,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88626921,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.622488498687744
    },
    {
      "auxiliary_loss_clip": 0.01168534,
      "auxiliary_loss_mlp": 0.01031239,
      "balance_loss_clip": 0.89160669,
      "balance_loss_mlp": 1.02390778,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 2.21405812876121,
      "language_loss": 0.74764299,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.76964074,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.8017852306365967
    },
    {
      "auxiliary_loss_clip": 0.01158456,
      "auxiliary_loss_mlp": 0.01122244,
      "balance_loss_clip": 0.89033401,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 2.1567563113353185,
      "language_loss": 0.80178624,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.82459331,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 4.518073081970215
    },
    {
      "auxiliary_loss_clip": 0.01170131,
      "auxiliary_loss_mlp": 0.01026745,
      "balance_loss_clip": 1.00800061,
      "balance_loss_mlp": 1.01931238,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 1.6950261027578366,
      "language_loss": 0.69893676,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.72090554,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.706909656524658
    },
    {
      "auxiliary_loss_clip": 0.01164274,
      "auxiliary_loss_mlp": 0.01028178,
      "balance_loss_clip": 0.97457379,
      "balance_loss_mlp": 1.02124333,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.9443066022455595,
      "language_loss": 0.73375028,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75567484,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 3.650038480758667
    },
    {
      "auxiliary_loss_clip": 0.01171799,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.01206374,
      "balance_loss_mlp": 1.02100432,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.4348826826703034,
      "language_loss": 0.80726641,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.82926732,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.6579971313476562
    },
    {
      "auxiliary_loss_clip": 0.01153815,
      "auxiliary_loss_mlp": 0.01024099,
      "balance_loss_clip": 0.8929615,
      "balance_loss_mlp": 1.01717281,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 1.8038794941928031,
      "language_loss": 0.72437739,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74615657,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.7397866249084473
    },
    {
      "auxiliary_loss_clip": 0.0116635,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.01236522,
      "balance_loss_mlp": 1.02002931,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.540358123768213,
      "language_loss": 0.68370956,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.7056433,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.6176626682281494
    },
    {
      "auxiliary_loss_clip": 0.01169972,
      "auxiliary_loss_mlp": 0.0102966,
      "balance_loss_clip": 0.97139692,
      "balance_loss_mlp": 1.02173257,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 1.8317359467019647,
      "language_loss": 0.81247914,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83447552,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.622565269470215
    },
    {
      "auxiliary_loss_clip": 0.01161529,
      "auxiliary_loss_mlp": 0.01028779,
      "balance_loss_clip": 0.97003269,
      "balance_loss_mlp": 1.02182627,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 1.8822539619852994,
      "language_loss": 0.84411317,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86601621,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.6796607971191406
    },
    {
      "auxiliary_loss_clip": 0.01159153,
      "auxiliary_loss_mlp": 0.01025221,
      "balance_loss_clip": 1.0063622,
      "balance_loss_mlp": 1.01786017,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 1.9148862507791835,
      "language_loss": 0.72425449,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74609828,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 2.722623825073242
    },
    {
      "auxiliary_loss_clip": 0.01079216,
      "auxiliary_loss_mlp": 0.01115764,
      "balance_loss_clip": 0.86301875,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.935515931459557,
      "language_loss": 0.64517975,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66712952,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 3.097550868988037
    },
    {
      "auxiliary_loss_clip": 0.01071957,
      "auxiliary_loss_mlp": 0.01115979,
      "balance_loss_clip": 0.90029734,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.085767660830572,
      "language_loss": 0.59686995,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61874932,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.204941987991333
    },
    {
      "auxiliary_loss_clip": 0.01169979,
      "auxiliary_loss_mlp": 0.0102647,
      "balance_loss_clip": 0.93277866,
      "balance_loss_mlp": 1.01884067,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.115970853759081,
      "language_loss": 0.76202112,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.78398561,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.6718366146087646
    },
    {
      "auxiliary_loss_clip": 0.01152532,
      "auxiliary_loss_mlp": 0.01026106,
      "balance_loss_clip": 0.81634372,
      "balance_loss_mlp": 1.01968026,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 1.7476926373665855,
      "language_loss": 0.6925385,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71432495,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 2.8082377910614014
    },
    {
      "auxiliary_loss_clip": 0.01167493,
      "auxiliary_loss_mlp": 0.01122138,
      "balance_loss_clip": 1.01102257,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 3.163929320701825,
      "language_loss": 0.75091851,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.7738148,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 2.6720197200775146
    },
    {
      "auxiliary_loss_clip": 0.01165385,
      "auxiliary_loss_mlp": 0.01022751,
      "balance_loss_clip": 0.97351193,
      "balance_loss_mlp": 1.01506829,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.65918527190369,
      "language_loss": 0.86577874,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.88766015,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.7206859588623047
    },
    {
      "auxiliary_loss_clip": 0.01156508,
      "auxiliary_loss_mlp": 0.01025076,
      "balance_loss_clip": 0.93168652,
      "balance_loss_mlp": 1.01793492,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.3455103394379122,
      "language_loss": 0.80104947,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82286531,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 2.7872672080993652
    },
    {
      "auxiliary_loss_clip": 0.01169131,
      "auxiliary_loss_mlp": 0.0102323,
      "balance_loss_clip": 0.97125596,
      "balance_loss_mlp": 1.01586282,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 1.6683973118821034,
      "language_loss": 0.80448782,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.82641137,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.643049955368042
    },
    {
      "auxiliary_loss_clip": 0.01154858,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 0.96881723,
      "balance_loss_mlp": 1.02159631,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.6141949863631502,
      "language_loss": 0.81188482,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.83371639,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 2.7132556438446045
    },
    {
      "auxiliary_loss_clip": 0.01162715,
      "auxiliary_loss_mlp": 0.0102852,
      "balance_loss_clip": 0.93058866,
      "balance_loss_mlp": 1.02094436,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.781402946058838,
      "language_loss": 0.86670911,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88862145,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.754638910293579
    },
    {
      "auxiliary_loss_clip": 0.01171114,
      "auxiliary_loss_mlp": 0.0102041,
      "balance_loss_clip": 0.93082356,
      "balance_loss_mlp": 1.01288223,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.7003678910995266,
      "language_loss": 0.8575213,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.87943649,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.7330310344696045
    },
    {
      "auxiliary_loss_clip": 0.01169444,
      "auxiliary_loss_mlp": 0.01122406,
      "balance_loss_clip": 1.04794955,
      "balance_loss_mlp": 0.0,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 1.6623229351723907,
      "language_loss": 0.76116157,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.78408003,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.628164529800415
    },
    {
      "auxiliary_loss_clip": 0.01171819,
      "auxiliary_loss_mlp": 0.0102798,
      "balance_loss_clip": 0.97316211,
      "balance_loss_mlp": 1.0203743,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 3.205451284228985,
      "language_loss": 0.85874933,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88074732,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 3.603203535079956
    },
    {
      "auxiliary_loss_clip": 0.01169553,
      "auxiliary_loss_mlp": 0.01031134,
      "balance_loss_clip": 1.01260889,
      "balance_loss_mlp": 1.0233736,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.156911418951469,
      "language_loss": 0.72357726,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.74558413,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.6548314094543457
    },
    {
      "auxiliary_loss_clip": 0.01164587,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 0.97209811,
      "balance_loss_mlp": 1.0214442,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 2.718179929792089,
      "language_loss": 0.49382079,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.51575458,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.665391445159912
    },
    {
      "auxiliary_loss_clip": 0.01162757,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 0.97311831,
      "balance_loss_mlp": 1.01932573,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 1.8767111379614996,
      "language_loss": 0.56224871,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.5841437,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 2.78193998336792
    },
    {
      "auxiliary_loss_clip": 0.01155633,
      "auxiliary_loss_mlp": 0.01027819,
      "balance_loss_clip": 1.00699091,
      "balance_loss_mlp": 1.02074385,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 4.524345279938104,
      "language_loss": 0.65744621,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.67928076,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 3.629969596862793
    },
    {
      "auxiliary_loss_clip": 0.01167206,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.01038361,
      "balance_loss_mlp": 1.02120018,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 2.7982657737578167,
      "language_loss": 0.71644723,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.73840654,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 3.621274471282959
    },
    {
      "auxiliary_loss_clip": 0.01162769,
      "auxiliary_loss_mlp": 0.01022813,
      "balance_loss_clip": 0.89429075,
      "balance_loss_mlp": 1.01514196,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.5667734707791454,
      "language_loss": 0.78149533,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80335116,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.782222270965576
    },
    {
      "auxiliary_loss_clip": 0.01163483,
      "auxiliary_loss_mlp": 0.01026465,
      "balance_loss_clip": 0.97110963,
      "balance_loss_mlp": 1.01958346,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 1.6279344757304595,
      "language_loss": 0.738657,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.76055646,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 3.5173304080963135
    },
    {
      "auxiliary_loss_clip": 0.01163179,
      "auxiliary_loss_mlp": 0.0102772,
      "balance_loss_clip": 0.89208353,
      "balance_loss_mlp": 1.02001929,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 1.8418705431568707,
      "language_loss": 0.73517239,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75708139,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.796205520629883
    },
    {
      "auxiliary_loss_clip": 0.01168286,
      "auxiliary_loss_mlp": 0.0102194,
      "balance_loss_clip": 0.97037923,
      "balance_loss_mlp": 1.01481676,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 1.9318826922813153,
      "language_loss": 0.70618647,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.72808874,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.6403627395629883
    },
    {
      "auxiliary_loss_clip": 0.01171151,
      "auxiliary_loss_mlp": 0.01024248,
      "balance_loss_clip": 1.0108273,
      "balance_loss_mlp": 1.01685739,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 17.73530751890689,
      "language_loss": 0.70238048,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72433448,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.6496918201446533
    },
    {
      "auxiliary_loss_clip": 0.01083252,
      "auxiliary_loss_mlp": 0.01001678,
      "balance_loss_clip": 0.91284895,
      "balance_loss_mlp": 1.00000942,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.9015032307224569,
      "language_loss": 0.60503775,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.6258871,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.191448926925659
    },
    {
      "auxiliary_loss_clip": 0.01171471,
      "auxiliary_loss_mlp": 0.01024489,
      "balance_loss_clip": 1.04944515,
      "balance_loss_mlp": 1.0174675,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 1.6568114863693415,
      "language_loss": 0.82396358,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84592319,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.631803035736084
    },
    {
      "auxiliary_loss_clip": 0.01168247,
      "auxiliary_loss_mlp": 0.01025798,
      "balance_loss_clip": 1.00834,
      "balance_loss_mlp": 1.01868737,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 1.8226103306728676,
      "language_loss": 0.82967556,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85161602,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 2.6637682914733887
    },
    {
      "auxiliary_loss_clip": 0.01079536,
      "auxiliary_loss_mlp": 0.01000025,
      "balance_loss_clip": 0.91104621,
      "balance_loss_mlp": 0.99817717,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 2.8580921840581617,
      "language_loss": 0.58195055,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.60274613,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.2942259311676025
    },
    {
      "auxiliary_loss_clip": 0.01169969,
      "auxiliary_loss_mlp": 0.01029194,
      "balance_loss_clip": 1.05072892,
      "balance_loss_mlp": 1.02242041,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 5.2473460841419195,
      "language_loss": 0.7667321,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.78872371,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 2.548375368118286
    },
    {
      "auxiliary_loss_clip": 0.01161362,
      "auxiliary_loss_mlp": 0.01122315,
      "balance_loss_clip": 0.92980063,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 1.6665534705568081,
      "language_loss": 0.81944513,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.84228194,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.725569725036621
    },
    {
      "auxiliary_loss_clip": 0.01170523,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 1.01176834,
      "balance_loss_mlp": 1.01738775,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.5461537581313347,
      "language_loss": 0.70893013,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.73088014,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.6854207515716553
    },
    {
      "auxiliary_loss_clip": 0.01158039,
      "auxiliary_loss_mlp": 0.01022706,
      "balance_loss_clip": 0.97055382,
      "balance_loss_mlp": 1.01618862,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 1.737144704576542,
      "language_loss": 0.77645373,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.79826123,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 2.6836540699005127
    },
    {
      "auxiliary_loss_clip": 0.01170306,
      "auxiliary_loss_mlp": 0.01022165,
      "balance_loss_clip": 0.89456713,
      "balance_loss_mlp": 1.01413012,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 2.0374895514347733,
      "language_loss": 0.81659919,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83852386,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 2.7543423175811768
    },
    {
      "auxiliary_loss_clip": 0.01170351,
      "auxiliary_loss_mlp": 0.01031195,
      "balance_loss_clip": 1.01194143,
      "balance_loss_mlp": 1.02409005,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 1.8965912775378395,
      "language_loss": 0.88344693,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.90546238,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 2.5746066570281982
    },
    {
      "auxiliary_loss_clip": 0.01152172,
      "auxiliary_loss_mlp": 0.01022168,
      "balance_loss_clip": 0.92901933,
      "balance_loss_mlp": 1.0149498,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.030769897992797,
      "language_loss": 0.72377169,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.74551511,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.7537941932678223
    },
    {
      "auxiliary_loss_clip": 0.01162227,
      "auxiliary_loss_mlp": 0.0102971,
      "balance_loss_clip": 0.96833622,
      "balance_loss_mlp": 1.02222323,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 3.8461147798198385,
      "language_loss": 0.62606966,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.64798903,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 2.7603492736816406
    },
    {
      "auxiliary_loss_clip": 0.01165833,
      "auxiliary_loss_mlp": 0.0102796,
      "balance_loss_clip": 0.93409336,
      "balance_loss_mlp": 1.02059317,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.2890415108303355,
      "language_loss": 0.76118684,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.7831248,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.791839838027954
    },
    {
      "auxiliary_loss_clip": 0.01176618,
      "auxiliary_loss_mlp": 0.01030246,
      "balance_loss_clip": 1.01173341,
      "balance_loss_mlp": 1.02232456,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.9148200888894609,
      "language_loss": 0.86928153,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.89135015,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.6560676097869873
    },
    {
      "auxiliary_loss_clip": 0.01168752,
      "auxiliary_loss_mlp": 0.01027171,
      "balance_loss_clip": 1.0498184,
      "balance_loss_mlp": 1.01977396,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 1.6398022930382,
      "language_loss": 0.67612123,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.69808042,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.739746570587158
    },
    {
      "auxiliary_loss_clip": 0.01174642,
      "auxiliary_loss_mlp": 0.01030007,
      "balance_loss_clip": 0.89817584,
      "balance_loss_mlp": 1.02235413,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 2.176317518770801,
      "language_loss": 0.71693343,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.73897994,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.8136672973632812
    },
    {
      "auxiliary_loss_clip": 0.01178947,
      "auxiliary_loss_mlp": 0.0102846,
      "balance_loss_clip": 0.97593451,
      "balance_loss_mlp": 1.02037799,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.199827348974353,
      "language_loss": 0.71069831,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73277235,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 3.766467332839966
    },
    {
      "auxiliary_loss_clip": 0.01172134,
      "auxiliary_loss_mlp": 0.01122769,
      "balance_loss_clip": 0.9337396,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 1.9383174891003077,
      "language_loss": 0.81269586,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.8356449,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.741273880004883
    },
    {
      "auxiliary_loss_clip": 0.01174624,
      "auxiliary_loss_mlp": 0.01026557,
      "balance_loss_clip": 0.93472064,
      "balance_loss_mlp": 1.01930022,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 1.7537287527627907,
      "language_loss": 0.68463165,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.70664346,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.763558864593506
    },
    {
      "auxiliary_loss_clip": 0.01070248,
      "auxiliary_loss_mlp": 0.01000521,
      "balance_loss_clip": 0.97541308,
      "balance_loss_mlp": 0.99879247,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7597895531532591,
      "language_loss": 0.58185899,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60256672,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.3001089096069336
    },
    {
      "auxiliary_loss_clip": 0.01160858,
      "auxiliary_loss_mlp": 0.01029753,
      "balance_loss_clip": 0.97204649,
      "balance_loss_mlp": 1.0225116,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 2.0747931515641063,
      "language_loss": 0.78238159,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.80428767,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 4.530453443527222
    },
    {
      "auxiliary_loss_clip": 0.01167013,
      "auxiliary_loss_mlp": 0.01028806,
      "balance_loss_clip": 0.97367752,
      "balance_loss_mlp": 1.02117622,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 1.8621928269454882,
      "language_loss": 0.79955316,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82151139,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.7226130962371826
    },
    {
      "auxiliary_loss_clip": 0.01159157,
      "auxiliary_loss_mlp": 0.01037476,
      "balance_loss_clip": 0.93162566,
      "balance_loss_mlp": 1.02972698,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 2.6003106346660814,
      "language_loss": 0.79831773,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82028407,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 3.6237845420837402
    },
    {
      "auxiliary_loss_clip": 0.0115275,
      "auxiliary_loss_mlp": 0.01027373,
      "balance_loss_clip": 0.92927933,
      "balance_loss_mlp": 1.02033997,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 1.520117594898483,
      "language_loss": 0.76024461,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.78204584,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.7762324810028076
    },
    {
      "auxiliary_loss_clip": 0.01150907,
      "auxiliary_loss_mlp": 0.01023569,
      "balance_loss_clip": 0.81727463,
      "balance_loss_mlp": 1.01605892,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.7098208243517568,
      "language_loss": 0.75063616,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77238095,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.867347240447998
    },
    {
      "auxiliary_loss_clip": 0.01170461,
      "auxiliary_loss_mlp": 0.01024616,
      "balance_loss_clip": 1.04901505,
      "balance_loss_mlp": 1.0171144,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 2.3830241896635163,
      "language_loss": 0.77910483,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80105567,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.745037078857422
    },
    {
      "auxiliary_loss_clip": 0.01160978,
      "auxiliary_loss_mlp": 0.01026451,
      "balance_loss_clip": 0.93307996,
      "balance_loss_mlp": 1.01879787,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 1.9099272398684675,
      "language_loss": 0.73262143,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.75449574,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.777757167816162
    },
    {
      "auxiliary_loss_clip": 0.01157832,
      "auxiliary_loss_mlp": 0.010227,
      "balance_loss_clip": 0.96898454,
      "balance_loss_mlp": 1.01456153,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 2.0680393750721486,
      "language_loss": 0.66943479,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.69124019,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.910466194152832
    },
    {
      "auxiliary_loss_clip": 0.01163135,
      "auxiliary_loss_mlp": 0.01027061,
      "balance_loss_clip": 0.97177434,
      "balance_loss_mlp": 1.01942575,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 2.52782304837913,
      "language_loss": 0.71801198,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.739914,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 2.773172378540039
    },
    {
      "auxiliary_loss_clip": 0.01170713,
      "auxiliary_loss_mlp": 0.010245,
      "balance_loss_clip": 1.01205015,
      "balance_loss_mlp": 1.01666152,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 2.0785482694533877,
      "language_loss": 0.74388576,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76583791,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 2.6580841541290283
    },
    {
      "auxiliary_loss_clip": 0.01166035,
      "auxiliary_loss_mlp": 0.01024265,
      "balance_loss_clip": 1.00955307,
      "balance_loss_mlp": 1.01755643,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 1.7076088174835873,
      "language_loss": 0.77163363,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.7935366,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 2.6902177333831787
    },
    {
      "auxiliary_loss_clip": 0.0116518,
      "auxiliary_loss_mlp": 0.0102674,
      "balance_loss_clip": 0.93618518,
      "balance_loss_mlp": 1.01888371,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 2.7145642919564086,
      "language_loss": 0.59161079,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.61352998,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 2.8156886100769043
    },
    {
      "auxiliary_loss_clip": 0.01169868,
      "auxiliary_loss_mlp": 0.01027277,
      "balance_loss_clip": 0.96981394,
      "balance_loss_mlp": 1.02016664,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 2.363636002484077,
      "language_loss": 0.8257488,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84772027,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.757575511932373
    },
    {
      "auxiliary_loss_clip": 0.01169163,
      "auxiliary_loss_mlp": 0.01029139,
      "balance_loss_clip": 0.97297752,
      "balance_loss_mlp": 1.02207565,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 2.0629370217447667,
      "language_loss": 0.67814845,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.70013148,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 2.713216543197632
    },
    {
      "auxiliary_loss_clip": 0.01171267,
      "auxiliary_loss_mlp": 0.01122796,
      "balance_loss_clip": 0.93530321,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 1.9972853845031369,
      "language_loss": 0.74555284,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.76849341,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 2.7209255695343018
    },
    {
      "auxiliary_loss_clip": 0.01171657,
      "auxiliary_loss_mlp": 0.01025086,
      "balance_loss_clip": 0.97138011,
      "balance_loss_mlp": 1.01763594,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.6671184229087985,
      "language_loss": 0.67863899,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.70060635,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 2.875032424926758
    },
    {
      "auxiliary_loss_clip": 0.01074601,
      "auxiliary_loss_mlp": 0.01001743,
      "balance_loss_clip": 0.98565173,
      "balance_loss_mlp": 0.99999088,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.811355767350827,
      "language_loss": 0.65106207,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67182553,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.2126681804656982
    },
    {
      "auxiliary_loss_clip": 0.01167624,
      "auxiliary_loss_mlp": 0.01122289,
      "balance_loss_clip": 0.94031549,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 1.5895219152372364,
      "language_loss": 0.75173891,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.77463806,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.7716829776763916
    },
    {
      "auxiliary_loss_clip": 0.01167558,
      "auxiliary_loss_mlp": 0.01023828,
      "balance_loss_clip": 0.97204322,
      "balance_loss_mlp": 1.01686358,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 1.6951292649894523,
      "language_loss": 0.83407617,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.85598999,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.7188656330108643
    },
    {
      "auxiliary_loss_clip": 0.01160378,
      "auxiliary_loss_mlp": 0.0102708,
      "balance_loss_clip": 0.9366039,
      "balance_loss_mlp": 1.02034199,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.7522480113784797,
      "language_loss": 0.71798229,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.73985684,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 2.8444008827209473
    },
    {
      "auxiliary_loss_clip": 0.01170412,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 0.93542862,
      "balance_loss_mlp": 1.01922214,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.8800549691615456,
      "language_loss": 0.73894268,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.76091409,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.798762321472168
    },
    {
      "auxiliary_loss_clip": 0.01168078,
      "auxiliary_loss_mlp": 0.01022876,
      "balance_loss_clip": 1.01013362,
      "balance_loss_mlp": 1.01527643,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.5349937290727658,
      "language_loss": 0.7038402,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.72574973,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 4.11737585067749
    },
    {
      "auxiliary_loss_clip": 0.01170391,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.01056349,
      "balance_loss_mlp": 1.01791525,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 1.6195112153074525,
      "language_loss": 0.73484409,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.7568,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.6721978187561035
    },
    {
      "auxiliary_loss_clip": 0.01149546,
      "auxiliary_loss_mlp": 0.01029821,
      "balance_loss_clip": 0.89112687,
      "balance_loss_mlp": 1.02199805,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.7457195823933112,
      "language_loss": 0.73380363,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75559723,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 2.737694263458252
    },
    {
      "auxiliary_loss_clip": 0.01172157,
      "auxiliary_loss_mlp": 0.01024396,
      "balance_loss_clip": 0.93268204,
      "balance_loss_mlp": 1.01683259,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 3.901674117379048,
      "language_loss": 0.61451477,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.63648027,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.7029805183410645
    },
    {
      "auxiliary_loss_clip": 0.01066312,
      "auxiliary_loss_mlp": 0.01000251,
      "balance_loss_clip": 0.90096968,
      "balance_loss_mlp": 0.99852288,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.801965599563055,
      "language_loss": 0.57528311,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.5959487,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.2499446868896484
    },
    {
      "auxiliary_loss_clip": 0.01170468,
      "auxiliary_loss_mlp": 0.01122555,
      "balance_loss_clip": 0.97265673,
      "balance_loss_mlp": 0.0,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 1.7481489100682177,
      "language_loss": 0.78652358,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.80945385,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 4.515204429626465
    },
    {
      "auxiliary_loss_clip": 0.01173505,
      "auxiliary_loss_mlp": 0.01025435,
      "balance_loss_clip": 1.0118897,
      "balance_loss_mlp": 1.01790738,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 1.8354023965089976,
      "language_loss": 0.61604917,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63803852,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.73683762550354
    },
    {
      "auxiliary_loss_clip": 0.01170337,
      "auxiliary_loss_mlp": 0.01029806,
      "balance_loss_clip": 0.89782929,
      "balance_loss_mlp": 1.02198601,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 1.862798685177662,
      "language_loss": 0.77835375,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.8003552,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 3.6836063861846924
    },
    {
      "auxiliary_loss_clip": 0.01169629,
      "auxiliary_loss_mlp": 0.01025182,
      "balance_loss_clip": 0.97197092,
      "balance_loss_mlp": 1.01804686,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.7487339534153457,
      "language_loss": 0.84920943,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.87115753,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 2.6017565727233887
    },
    {
      "auxiliary_loss_clip": 0.01152376,
      "auxiliary_loss_mlp": 0.01029924,
      "balance_loss_clip": 0.81707525,
      "balance_loss_mlp": 1.02237773,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 1.9190281423341033,
      "language_loss": 0.68570423,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.70752722,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.9503934383392334
    },
    {
      "auxiliary_loss_clip": 0.01165152,
      "auxiliary_loss_mlp": 0.01029411,
      "balance_loss_clip": 0.97391212,
      "balance_loss_mlp": 1.02188039,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 1.6368960755240827,
      "language_loss": 0.71832603,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74027169,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 3.331721782684326
    },
    {
      "auxiliary_loss_clip": 0.01163638,
      "auxiliary_loss_mlp": 0.01025649,
      "balance_loss_clip": 0.93283111,
      "balance_loss_mlp": 1.01816869,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 1.9659074623360773,
      "language_loss": 0.82335365,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84524655,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.7123372554779053
    },
    {
      "auxiliary_loss_clip": 0.01162374,
      "auxiliary_loss_mlp": 0.0112258,
      "balance_loss_clip": 0.93123996,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 1.8773877304124826,
      "language_loss": 0.90312028,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.92596984,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.771224021911621
    },
    {
      "auxiliary_loss_clip": 0.01160985,
      "auxiliary_loss_mlp": 0.01028152,
      "balance_loss_clip": 0.96926868,
      "balance_loss_mlp": 1.02042139,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.6345749514895238,
      "language_loss": 0.72145152,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74334288,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.8455045223236084
    },
    {
      "auxiliary_loss_clip": 0.0116493,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 1.00774372,
      "balance_loss_mlp": 1.02013183,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 2.0730178927887164,
      "language_loss": 0.77163625,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.79355729,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 2.8178703784942627
    },
    {
      "auxiliary_loss_clip": 0.01167635,
      "auxiliary_loss_mlp": 0.01022386,
      "balance_loss_clip": 1.01034176,
      "balance_loss_mlp": 1.01597297,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 1.682317274905178,
      "language_loss": 0.84911281,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87101305,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 2.588503122329712
    },
    {
      "auxiliary_loss_clip": 0.01161599,
      "auxiliary_loss_mlp": 0.01028323,
      "balance_loss_clip": 0.93325675,
      "balance_loss_mlp": 1.02115238,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 1.6679339379203766,
      "language_loss": 0.77927679,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80117595,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.732787609100342
    },
    {
      "auxiliary_loss_clip": 0.01169834,
      "auxiliary_loss_mlp": 0.01022603,
      "balance_loss_clip": 0.93547869,
      "balance_loss_mlp": 1.01541758,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.4799506256888135,
      "language_loss": 0.80734372,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.8292681,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 2.7898004055023193
    },
    {
      "auxiliary_loss_clip": 0.01169983,
      "auxiliary_loss_mlp": 0.01022711,
      "balance_loss_clip": 1.04910672,
      "balance_loss_mlp": 1.01545763,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 1.7871907515671308,
      "language_loss": 0.77689743,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.79882437,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.6556529998779297
    },
    {
      "auxiliary_loss_clip": 0.0116974,
      "auxiliary_loss_mlp": 0.01022553,
      "balance_loss_clip": 0.93553305,
      "balance_loss_mlp": 1.01487637,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 1.8843622766123442,
      "language_loss": 0.87118983,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.89311278,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 2.760502815246582
    },
    {
      "auxiliary_loss_clip": 0.01175317,
      "auxiliary_loss_mlp": 0.01025816,
      "balance_loss_clip": 0.89584708,
      "balance_loss_mlp": 1.0183115,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 1.9032098374654483,
      "language_loss": 0.88395107,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.90596241,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.765763282775879
    },
    {
      "auxiliary_loss_clip": 0.01173517,
      "auxiliary_loss_mlp": 0.01026762,
      "balance_loss_clip": 0.97461677,
      "balance_loss_mlp": 1.01983309,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 1.7992916936425494,
      "language_loss": 0.72819865,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.75020146,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.677875280380249
    },
    {
      "auxiliary_loss_clip": 0.01166354,
      "auxiliary_loss_mlp": 0.01024944,
      "balance_loss_clip": 0.9732703,
      "balance_loss_mlp": 1.01760364,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 2.2926157496203396,
      "language_loss": 0.68215829,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.70407128,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 2.668498992919922
    },
    {
      "auxiliary_loss_clip": 0.01167466,
      "auxiliary_loss_mlp": 0.01025112,
      "balance_loss_clip": 0.89759851,
      "balance_loss_mlp": 1.01740849,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.655924117856182,
      "language_loss": 0.74697387,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.76889968,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 2.759160041809082
    },
    {
      "auxiliary_loss_clip": 0.01158664,
      "auxiliary_loss_mlp": 0.01029924,
      "balance_loss_clip": 0.89558375,
      "balance_loss_mlp": 1.02312589,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.4744829970807902,
      "language_loss": 0.74882215,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77070808,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.78658390045166
    },
    {
      "auxiliary_loss_clip": 0.01163226,
      "auxiliary_loss_mlp": 0.0103021,
      "balance_loss_clip": 0.97224551,
      "balance_loss_mlp": 1.02273524,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 1.814517599962057,
      "language_loss": 0.7466563,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.76859069,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.699347734451294
    },
    {
      "auxiliary_loss_clip": 0.01168429,
      "auxiliary_loss_mlp": 0.01027473,
      "balance_loss_clip": 1.04841423,
      "balance_loss_mlp": 1.02016568,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 1.9765285149551843,
      "language_loss": 0.72504735,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.74700642,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 3.9878687858581543
    },
    {
      "auxiliary_loss_clip": 0.01175972,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.01099098,
      "balance_loss_mlp": 1.02064013,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.660889943664633,
      "language_loss": 0.72362399,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.7456755,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.657158136367798
    },
    {
      "auxiliary_loss_clip": 0.01168875,
      "auxiliary_loss_mlp": 0.01021403,
      "balance_loss_clip": 1.0492568,
      "balance_loss_mlp": 1.01376784,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.5350143052113252,
      "language_loss": 0.68727589,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.70917863,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 2.795463800430298
    },
    {
      "auxiliary_loss_clip": 0.01172559,
      "auxiliary_loss_mlp": 0.01026578,
      "balance_loss_clip": 1.05072331,
      "balance_loss_mlp": 1.01881146,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 1.9425069010725218,
      "language_loss": 0.71787226,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.73986369,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.663937568664551
    },
    {
      "auxiliary_loss_clip": 0.011697,
      "auxiliary_loss_mlp": 0.01122585,
      "balance_loss_clip": 1.04805422,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.2892567492948794,
      "language_loss": 0.76663792,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.78956079,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.6661643981933594
    },
    {
      "auxiliary_loss_clip": 0.01173622,
      "auxiliary_loss_mlp": 0.01122681,
      "balance_loss_clip": 0.97240007,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 2.025909740540363,
      "language_loss": 0.6916818,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.71464479,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 3.676208019256592
    },
    {
      "auxiliary_loss_clip": 0.01161919,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.00926948,
      "balance_loss_mlp": 1.02227151,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 2.0777219134332983,
      "language_loss": 0.79602611,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.81793952,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.7439322471618652
    },
    {
      "auxiliary_loss_clip": 0.01168916,
      "auxiliary_loss_mlp": 0.01027376,
      "balance_loss_clip": 1.04806137,
      "balance_loss_mlp": 1.02032495,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 1.6506348115523695,
      "language_loss": 0.7749809,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.79694378,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 3.607912063598633
    },
    {
      "auxiliary_loss_clip": 0.01171957,
      "auxiliary_loss_mlp": 0.01123121,
      "balance_loss_clip": 1.01078486,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 2.145643394939609,
      "language_loss": 0.6807611,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.70371193,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.7561984062194824
    },
    {
      "auxiliary_loss_clip": 0.01159937,
      "auxiliary_loss_mlp": 0.01032268,
      "balance_loss_clip": 0.93168652,
      "balance_loss_mlp": 1.02496636,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 2.689521198285791,
      "language_loss": 0.77462268,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79654479,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.801934242248535
    },
    {
      "auxiliary_loss_clip": 0.01170315,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 0.97205377,
      "balance_loss_mlp": 1.01751339,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.5685028175794336,
      "language_loss": 0.74441385,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.7663638,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.7331936359405518
    },
    {
      "auxiliary_loss_clip": 0.01163315,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 0.96890205,
      "balance_loss_mlp": 1.01731932,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 3.034012393970634,
      "language_loss": 0.8506121,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87249476,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.683784008026123
    },
    {
      "auxiliary_loss_clip": 0.0116797,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.04944825,
      "balance_loss_mlp": 1.01839399,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.5584607435082374,
      "language_loss": 0.82241917,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84435165,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.6760189533233643
    },
    {
      "auxiliary_loss_clip": 0.01169736,
      "auxiliary_loss_mlp": 0.01027885,
      "balance_loss_clip": 1.00907433,
      "balance_loss_mlp": 1.02002954,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 2.5070392909777155,
      "language_loss": 0.62180853,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.6437847,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.724057912826538
    },
    {
      "auxiliary_loss_clip": 0.01173519,
      "auxiliary_loss_mlp": 0.01023035,
      "balance_loss_clip": 0.9730041,
      "balance_loss_mlp": 1.01629925,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 2.187639015542998,
      "language_loss": 0.81110644,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.83307195,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 2.745480537414551
    },
    {
      "auxiliary_loss_clip": 0.01166421,
      "auxiliary_loss_mlp": 0.01025856,
      "balance_loss_clip": 1.00844026,
      "balance_loss_mlp": 1.01886404,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 1.5790654727108413,
      "language_loss": 0.78095275,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80287552,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 2.649782180786133
    },
    {
      "auxiliary_loss_clip": 0.01172057,
      "auxiliary_loss_mlp": 0.0103223,
      "balance_loss_clip": 1.01091373,
      "balance_loss_mlp": 1.02399886,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 3.513001711586992,
      "language_loss": 0.87948304,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.90152586,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.731992483139038
    },
    {
      "auxiliary_loss_clip": 0.01171112,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 0.93617135,
      "balance_loss_mlp": 1.02030754,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 5.462495598555303,
      "language_loss": 0.71908879,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.7410751,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 2.7184319496154785
    },
    {
      "auxiliary_loss_clip": 0.01170583,
      "auxiliary_loss_mlp": 0.01028016,
      "balance_loss_clip": 1.01018918,
      "balance_loss_mlp": 1.02091742,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 2.7448626316167584,
      "language_loss": 0.62382698,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.64581299,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 2.738820791244507
    },
    {
      "auxiliary_loss_clip": 0.01168005,
      "auxiliary_loss_mlp": 0.01020864,
      "balance_loss_clip": 1.00951457,
      "balance_loss_mlp": 1.01408362,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 1.6976928373206681,
      "language_loss": 0.74524295,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.76713163,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 2.7081236839294434
    },
    {
      "auxiliary_loss_clip": 0.01154504,
      "auxiliary_loss_mlp": 0.0102543,
      "balance_loss_clip": 0.96614182,
      "balance_loss_mlp": 1.01820576,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 1.7037546602567757,
      "language_loss": 0.65944481,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68124413,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 2.7918732166290283
    },
    {
      "auxiliary_loss_clip": 0.01171397,
      "auxiliary_loss_mlp": 0.01027558,
      "balance_loss_clip": 1.01385212,
      "balance_loss_mlp": 1.01957726,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 1.7689855845615325,
      "language_loss": 0.70280927,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.7247988,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.755159378051758
    },
    {
      "auxiliary_loss_clip": 0.01161524,
      "auxiliary_loss_mlp": 0.01023924,
      "balance_loss_clip": 1.00952506,
      "balance_loss_mlp": 1.01643753,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 1.9771743005349534,
      "language_loss": 0.62758785,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64944232,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 2.914186954498291
    },
    {
      "auxiliary_loss_clip": 0.01152196,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 0.93055141,
      "balance_loss_mlp": 1.01999962,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 1.7873809210324887,
      "language_loss": 0.70680749,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.72860944,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.7802789211273193
    },
    {
      "auxiliary_loss_clip": 0.01157827,
      "auxiliary_loss_mlp": 0.01027278,
      "balance_loss_clip": 0.89454395,
      "balance_loss_mlp": 1.01942801,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.3382282968513888,
      "language_loss": 0.83238232,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85423338,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.7772727012634277
    },
    {
      "auxiliary_loss_clip": 0.01167061,
      "auxiliary_loss_mlp": 0.01026764,
      "balance_loss_clip": 0.97271764,
      "balance_loss_mlp": 1.01923609,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.8747326301563458,
      "language_loss": 0.8991375,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.9210757,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.7690422534942627
    },
    {
      "auxiliary_loss_clip": 0.0117075,
      "auxiliary_loss_mlp": 0.01024325,
      "balance_loss_clip": 1.05060518,
      "balance_loss_mlp": 1.01716614,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.7284915461247061,
      "language_loss": 0.76736367,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.78931445,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 3.7203803062438965
    },
    {
      "auxiliary_loss_clip": 0.01170686,
      "auxiliary_loss_mlp": 0.0102229,
      "balance_loss_clip": 0.93576336,
      "balance_loss_mlp": 1.01486301,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.5769430957799797,
      "language_loss": 0.72298801,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.74491775,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.747521162033081
    },
    {
      "auxiliary_loss_clip": 0.01167942,
      "auxiliary_loss_mlp": 0.01021699,
      "balance_loss_clip": 1.01117539,
      "balance_loss_mlp": 1.01451683,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 1.9814822203087878,
      "language_loss": 0.75056559,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.77246201,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 2.655562162399292
    },
    {
      "auxiliary_loss_clip": 0.01166281,
      "auxiliary_loss_mlp": 0.01032236,
      "balance_loss_clip": 0.89784455,
      "balance_loss_mlp": 1.0243355,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 1.8119451016267085,
      "language_loss": 0.69461071,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71659595,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 2.7394304275512695
    },
    {
      "auxiliary_loss_clip": 0.01166566,
      "auxiliary_loss_mlp": 0.01030418,
      "balance_loss_clip": 1.00844479,
      "balance_loss_mlp": 1.02340865,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 1.8208760136819842,
      "language_loss": 0.7771672,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.799137,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.7038917541503906
    },
    {
      "auxiliary_loss_clip": 0.01169462,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 0.89885515,
      "balance_loss_mlp": 1.02020729,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 1.7163092813389371,
      "language_loss": 0.78767419,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.80964243,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 4.672168493270874
    },
    {
      "auxiliary_loss_clip": 0.01169904,
      "auxiliary_loss_mlp": 0.01024563,
      "balance_loss_clip": 1.04912031,
      "balance_loss_mlp": 1.01708305,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 1.6528001220865525,
      "language_loss": 0.78484732,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.80679196,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 3.650085687637329
    },
    {
      "auxiliary_loss_clip": 0.01159257,
      "auxiliary_loss_mlp": 0.01022376,
      "balance_loss_clip": 0.93258941,
      "balance_loss_mlp": 1.01487219,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 1.8773690637748153,
      "language_loss": 0.76953304,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.79134935,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.8758318424224854
    },
    {
      "auxiliary_loss_clip": 0.01173529,
      "auxiliary_loss_mlp": 0.01024795,
      "balance_loss_clip": 0.97420192,
      "balance_loss_mlp": 1.01710284,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 1.9733324500359564,
      "language_loss": 0.81428021,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.83626348,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.8122901916503906
    },
    {
      "auxiliary_loss_clip": 0.01156187,
      "auxiliary_loss_mlp": 0.01023408,
      "balance_loss_clip": 0.96803391,
      "balance_loss_mlp": 1.01585603,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 1.5921270360915583,
      "language_loss": 0.76295853,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.78475451,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.8518550395965576
    },
    {
      "auxiliary_loss_clip": 0.01167808,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 0.97254449,
      "balance_loss_mlp": 1.01750088,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 1.9402891358145757,
      "language_loss": 0.67830265,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70022297,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.7928221225738525
    },
    {
      "auxiliary_loss_clip": 0.0116971,
      "auxiliary_loss_mlp": 0.01029124,
      "balance_loss_clip": 0.93204033,
      "balance_loss_mlp": 1.02125025,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 2.787149205195021,
      "language_loss": 0.66106212,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68305045,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.7997825145721436
    },
    {
      "auxiliary_loss_clip": 0.01160861,
      "auxiliary_loss_mlp": 0.01025295,
      "balance_loss_clip": 1.00888634,
      "balance_loss_mlp": 1.01818752,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.068225916488796,
      "language_loss": 0.75185335,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.7737149,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.7097907066345215
    },
    {
      "auxiliary_loss_clip": 0.01164892,
      "auxiliary_loss_mlp": 0.0102409,
      "balance_loss_clip": 0.97353703,
      "balance_loss_mlp": 1.01693201,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 2.5819366612550354,
      "language_loss": 0.73939043,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.7612803,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.749863624572754
    },
    {
      "auxiliary_loss_clip": 0.01168385,
      "auxiliary_loss_mlp": 0.01025369,
      "balance_loss_clip": 1.0112499,
      "balance_loss_mlp": 1.01796067,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 2.1965779570774635,
      "language_loss": 0.76598442,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.78792197,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 2.715283155441284
    },
    {
      "auxiliary_loss_clip": 0.01166557,
      "auxiliary_loss_mlp": 0.01025075,
      "balance_loss_clip": 1.04908776,
      "balance_loss_mlp": 1.01767862,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 1.9627057136179273,
      "language_loss": 0.79463816,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.81655443,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 2.6889004707336426
    },
    {
      "auxiliary_loss_clip": 0.01172816,
      "auxiliary_loss_mlp": 0.01032169,
      "balance_loss_clip": 1.05099821,
      "balance_loss_mlp": 1.02456975,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 2.7678832981017703,
      "language_loss": 0.79959613,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.82164598,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 2.6766233444213867
    },
    {
      "auxiliary_loss_clip": 0.01168886,
      "auxiliary_loss_mlp": 0.01026418,
      "balance_loss_clip": 1.04801941,
      "balance_loss_mlp": 1.01945066,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 1.7956837842529743,
      "language_loss": 0.56847012,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59042317,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.727919816970825
    },
    {
      "auxiliary_loss_clip": 0.01170345,
      "auxiliary_loss_mlp": 0.01027141,
      "balance_loss_clip": 0.89943302,
      "balance_loss_mlp": 1.01983929,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 2.106641083275422,
      "language_loss": 0.76990438,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.7918793,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 2.8625173568725586
    },
    {
      "auxiliary_loss_clip": 0.01172681,
      "auxiliary_loss_mlp": 0.01024734,
      "balance_loss_clip": 1.05019808,
      "balance_loss_mlp": 1.01736689,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 1.8860054068556782,
      "language_loss": 0.79151118,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.81348538,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 2.6450023651123047
    },
    {
      "auxiliary_loss_clip": 0.01170374,
      "auxiliary_loss_mlp": 0.01024391,
      "balance_loss_clip": 1.00994885,
      "balance_loss_mlp": 1.01694012,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 1.7072570542408736,
      "language_loss": 0.84631586,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86826354,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 2.693148374557495
    },
    {
      "auxiliary_loss_clip": 0.01165508,
      "auxiliary_loss_mlp": 0.01032666,
      "balance_loss_clip": 0.93038464,
      "balance_loss_mlp": 1.02513742,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 1.777636620277453,
      "language_loss": 0.73564231,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.75762409,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 3.0466885566711426
    },
    {
      "auxiliary_loss_clip": 0.01165383,
      "auxiliary_loss_mlp": 0.01026421,
      "balance_loss_clip": 0.93341577,
      "balance_loss_mlp": 1.0188396,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 2.4865692974504,
      "language_loss": 0.70355713,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.72547519,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 2.8327605724334717
    },
    {
      "auxiliary_loss_clip": 0.0116938,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 0.97227466,
      "balance_loss_mlp": 1.01736569,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 3.936021355987275,
      "language_loss": 0.75813252,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.78006947,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 2.855226755142212
    },
    {
      "auxiliary_loss_clip": 0.01164775,
      "auxiliary_loss_mlp": 0.01032372,
      "balance_loss_clip": 0.97274947,
      "balance_loss_mlp": 1.02507937,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 3.6867074097101873,
      "language_loss": 0.77710003,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.79907155,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.7521238327026367
    },
    {
      "auxiliary_loss_clip": 0.01162844,
      "auxiliary_loss_mlp": 0.01022354,
      "balance_loss_clip": 0.97263342,
      "balance_loss_mlp": 1.01532984,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 1.9647517364055966,
      "language_loss": 0.70688111,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72873306,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.751574754714966
    },
    {
      "auxiliary_loss_clip": 0.01164283,
      "auxiliary_loss_mlp": 0.01021689,
      "balance_loss_clip": 0.97258705,
      "balance_loss_mlp": 1.01471519,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.6867550395269963,
      "language_loss": 0.8206774,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84253711,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.7152600288391113
    },
    {
      "auxiliary_loss_clip": 0.01158889,
      "auxiliary_loss_mlp": 0.01023621,
      "balance_loss_clip": 0.8944056,
      "balance_loss_mlp": 1.01677811,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 2.2925130182669378,
      "language_loss": 0.80161989,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82344496,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 3.7003579139709473
    },
    {
      "auxiliary_loss_clip": 0.01159279,
      "auxiliary_loss_mlp": 0.01021991,
      "balance_loss_clip": 1.00783098,
      "balance_loss_mlp": 1.01413536,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 2.238386290509437,
      "language_loss": 0.7626195,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.78443217,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 2.7143948078155518
    },
    {
      "auxiliary_loss_clip": 0.01161542,
      "auxiliary_loss_mlp": 0.01028797,
      "balance_loss_clip": 1.00918794,
      "balance_loss_mlp": 1.02174616,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 2.3026742970753484,
      "language_loss": 0.76600701,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.7879104,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 2.698866128921509
    },
    {
      "auxiliary_loss_clip": 0.01171388,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.01229513,
      "balance_loss_mlp": 1.01856542,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 1.9779526216852799,
      "language_loss": 0.85917175,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.88114661,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.7037458419799805
    },
    {
      "auxiliary_loss_clip": 0.01149715,
      "auxiliary_loss_mlp": 0.01031144,
      "balance_loss_clip": 0.89284486,
      "balance_loss_mlp": 1.02376807,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 1.8150780997889286,
      "language_loss": 0.78504092,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.80684948,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 3.76906681060791
    },
    {
      "auxiliary_loss_clip": 0.01159659,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 0.93358111,
      "balance_loss_mlp": 1.01726139,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 1.7563958787941838,
      "language_loss": 0.75390697,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.7757498,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 4.671457052230835
    },
    {
      "auxiliary_loss_clip": 0.0116903,
      "auxiliary_loss_mlp": 0.0102716,
      "balance_loss_clip": 1.05017185,
      "balance_loss_mlp": 1.01993871,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 4.41895663099938,
      "language_loss": 0.70071769,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72267962,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.6851987838745117
    },
    {
      "auxiliary_loss_clip": 0.01166992,
      "auxiliary_loss_mlp": 0.01027836,
      "balance_loss_clip": 1.0128969,
      "balance_loss_mlp": 1.02103841,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.0509747945477668,
      "language_loss": 0.76894599,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.79089427,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.6576967239379883
    },
    {
      "auxiliary_loss_clip": 0.0117305,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 1.05007744,
      "balance_loss_mlp": 1.01937938,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.1041691476585913,
      "language_loss": 0.80788714,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.82988864,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.711019515991211
    },
    {
      "auxiliary_loss_clip": 0.0106903,
      "auxiliary_loss_mlp": 0.01004316,
      "balance_loss_clip": 0.97459078,
      "balance_loss_mlp": 1.00263524,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8109409424543004,
      "language_loss": 0.58277357,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.60350704,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.345893383026123
    },
    {
      "auxiliary_loss_clip": 0.01175111,
      "auxiliary_loss_mlp": 0.0102851,
      "balance_loss_clip": 0.89612919,
      "balance_loss_mlp": 1.02054691,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 1.9017442482434663,
      "language_loss": 0.79850614,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.82054234,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.8722829818725586
    },
    {
      "auxiliary_loss_clip": 0.01168158,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.01143551,
      "balance_loss_mlp": 1.01995397,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.6628508409681853,
      "language_loss": 0.91776419,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.93971777,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.8095815181732178
    },
    {
      "auxiliary_loss_clip": 0.01175331,
      "auxiliary_loss_mlp": 0.01122573,
      "balance_loss_clip": 0.97210944,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 1.6354240542548903,
      "language_loss": 0.84379637,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.86677539,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.7434375286102295
    },
    {
      "auxiliary_loss_clip": 0.01171718,
      "auxiliary_loss_mlp": 0.01123059,
      "balance_loss_clip": 1.04918671,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 1.9598862441927833,
      "language_loss": 0.69029748,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.71324527,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.718404531478882
    },
    {
      "auxiliary_loss_clip": 0.01171817,
      "auxiliary_loss_mlp": 0.01030102,
      "balance_loss_clip": 1.01005805,
      "balance_loss_mlp": 1.02260995,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 1.7250504098096784,
      "language_loss": 0.6994487,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72146791,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 2.703331470489502
    },
    {
      "auxiliary_loss_clip": 0.01169385,
      "auxiliary_loss_mlp": 0.01122571,
      "balance_loss_clip": 1.01267445,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 2.014023357231085,
      "language_loss": 0.80068433,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.82360387,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 2.71642804145813
    },
    {
      "auxiliary_loss_clip": 0.01159669,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 0.97219861,
      "balance_loss_mlp": 1.02107811,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 1.9065550777482276,
      "language_loss": 0.70956564,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.73144817,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 2.7525885105133057
    },
    {
      "auxiliary_loss_clip": 0.01170122,
      "auxiliary_loss_mlp": 0.01025858,
      "balance_loss_clip": 0.97550964,
      "balance_loss_mlp": 1.01869941,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 1.8334565250291326,
      "language_loss": 0.7867704,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.80873019,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 2.739002227783203
    },
    {
      "auxiliary_loss_clip": 0.01167184,
      "auxiliary_loss_mlp": 0.01025605,
      "balance_loss_clip": 1.0106194,
      "balance_loss_mlp": 1.01858354,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.6302350768898426,
      "language_loss": 0.77894545,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.80087334,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 2.676457405090332
    },
    {
      "auxiliary_loss_clip": 0.01169323,
      "auxiliary_loss_mlp": 0.01024128,
      "balance_loss_clip": 1.01094282,
      "balance_loss_mlp": 1.01718378,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 1.5887572220798925,
      "language_loss": 0.82508272,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.84701723,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 2.6660804748535156
    },
    {
      "auxiliary_loss_clip": 0.01168925,
      "auxiliary_loss_mlp": 0.01123138,
      "balance_loss_clip": 0.9740265,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 1.8767358192251002,
      "language_loss": 0.78490686,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.80782753,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.7508363723754883
    },
    {
      "auxiliary_loss_clip": 0.01164749,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 0.89629799,
      "balance_loss_mlp": 1.01880634,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 2.1323323666921614,
      "language_loss": 0.56836021,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.59027833,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 2.845510721206665
    },
    {
      "auxiliary_loss_clip": 0.01169804,
      "auxiliary_loss_mlp": 0.01030273,
      "balance_loss_clip": 1.01310921,
      "balance_loss_mlp": 1.02296507,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 1.7169024637788088,
      "language_loss": 0.74214137,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76414216,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 2.6878719329833984
    },
    {
      "auxiliary_loss_clip": 0.01173771,
      "auxiliary_loss_mlp": 0.01024293,
      "balance_loss_clip": 0.85560143,
      "balance_loss_mlp": 1.01697314,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 1.6827469666721449,
      "language_loss": 0.74168283,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.76366341,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.8942880630493164
    },
    {
      "auxiliary_loss_clip": 0.01176612,
      "auxiliary_loss_mlp": 0.01026973,
      "balance_loss_clip": 0.9759202,
      "balance_loss_mlp": 1.01905787,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 1.7075327381142746,
      "language_loss": 0.68934882,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71138465,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.82381010055542
    },
    {
      "auxiliary_loss_clip": 0.01170004,
      "auxiliary_loss_mlp": 0.01023591,
      "balance_loss_clip": 1.04866219,
      "balance_loss_mlp": 1.01667058,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 1.7748102385705673,
      "language_loss": 0.74281323,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76474917,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.6052701473236084
    },
    {
      "auxiliary_loss_clip": 0.01068863,
      "auxiliary_loss_mlp": 0.01002015,
      "balance_loss_clip": 0.9378944,
      "balance_loss_mlp": 1.00022686,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7752514441473022,
      "language_loss": 0.5575887,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57829744,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 4.2026286125183105
    },
    {
      "auxiliary_loss_clip": 0.01160409,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 1.0082109,
      "balance_loss_mlp": 1.01772332,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 1.5823477582079888,
      "language_loss": 0.67645508,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.69831228,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.6529088020324707
    },
    {
      "auxiliary_loss_clip": 0.01172127,
      "auxiliary_loss_mlp": 0.01025419,
      "balance_loss_clip": 1.0475657,
      "balance_loss_mlp": 1.01780748,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 1.9692109054014901,
      "language_loss": 0.81208956,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83406496,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 2.5994441509246826
    },
    {
      "auxiliary_loss_clip": 0.01160366,
      "auxiliary_loss_mlp": 0.01024029,
      "balance_loss_clip": 0.93278289,
      "balance_loss_mlp": 1.01726127,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 1.9054456173444487,
      "language_loss": 0.60513484,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.62697881,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.8194217681884766
    },
    {
      "auxiliary_loss_clip": 0.01163209,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 0.93251055,
      "balance_loss_mlp": 1.02101588,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.5965959803290617,
      "language_loss": 0.64829457,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.67020702,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 3.7803969383239746
    },
    {
      "auxiliary_loss_clip": 0.01165834,
      "auxiliary_loss_mlp": 0.01021134,
      "balance_loss_clip": 0.97311211,
      "balance_loss_mlp": 1.01396918,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.3117311586740414,
      "language_loss": 0.77444577,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.79631543,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 3.5756607055664062
    },
    {
      "auxiliary_loss_clip": 0.01153825,
      "auxiliary_loss_mlp": 0.0102548,
      "balance_loss_clip": 0.93092233,
      "balance_loss_mlp": 1.01813114,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 2.2254767943387552,
      "language_loss": 0.77244496,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.79423809,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 3.6369428634643555
    },
    {
      "auxiliary_loss_clip": 0.01159222,
      "auxiliary_loss_mlp": 0.01030127,
      "balance_loss_clip": 0.932473,
      "balance_loss_mlp": 1.02259922,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 2.1132131437987884,
      "language_loss": 0.79266846,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.81456196,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.7362940311431885
    },
    {
      "auxiliary_loss_clip": 0.01164135,
      "auxiliary_loss_mlp": 0.01024666,
      "balance_loss_clip": 1.00918627,
      "balance_loss_mlp": 1.01773095,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 1.830903629551356,
      "language_loss": 0.68592489,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70781291,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.714681625366211
    },
    {
      "auxiliary_loss_clip": 0.01168816,
      "auxiliary_loss_mlp": 0.01025691,
      "balance_loss_clip": 1.05068278,
      "balance_loss_mlp": 1.01813889,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 2.4408012104584835,
      "language_loss": 0.82094377,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84288883,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.6488146781921387
    },
    {
      "auxiliary_loss_clip": 0.01157474,
      "auxiliary_loss_mlp": 0.01028566,
      "balance_loss_clip": 0.89526492,
      "balance_loss_mlp": 1.02155697,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 3.1356667880553863,
      "language_loss": 0.78692281,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.80878317,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.8464202880859375
    },
    {
      "auxiliary_loss_clip": 0.01167875,
      "auxiliary_loss_mlp": 0.01023502,
      "balance_loss_clip": 1.04730272,
      "balance_loss_mlp": 1.0160929,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.7058054767902606,
      "language_loss": 0.80758494,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.82949871,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.711348533630371
    },
    {
      "auxiliary_loss_clip": 0.01154256,
      "auxiliary_loss_mlp": 0.01021857,
      "balance_loss_clip": 0.93120718,
      "balance_loss_mlp": 1.01407254,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 1.6630356142812075,
      "language_loss": 0.68754327,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.70930439,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 3.2046408653259277
    },
    {
      "auxiliary_loss_clip": 0.01168478,
      "auxiliary_loss_mlp": 0.01025318,
      "balance_loss_clip": 1.01260567,
      "balance_loss_mlp": 1.01816809,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 1.911178849658274,
      "language_loss": 0.80404222,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.82598019,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.7065742015838623
    },
    {
      "auxiliary_loss_clip": 0.01164533,
      "auxiliary_loss_mlp": 0.01022781,
      "balance_loss_clip": 0.93417442,
      "balance_loss_mlp": 1.01553297,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 2.292368359581844,
      "language_loss": 0.82582188,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.84769505,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 2.7575085163116455
    },
    {
      "auxiliary_loss_clip": 0.01163989,
      "auxiliary_loss_mlp": 0.01020373,
      "balance_loss_clip": 1.01283288,
      "balance_loss_mlp": 1.01353335,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.74484032963186,
      "language_loss": 0.78520823,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.8070519,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 2.7539103031158447
    },
    {
      "auxiliary_loss_clip": 0.01173271,
      "auxiliary_loss_mlp": 0.01025204,
      "balance_loss_clip": 0.93179661,
      "balance_loss_mlp": 1.01788449,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.398344750094204,
      "language_loss": 0.78635561,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.80834037,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 2.8056774139404297
    },
    {
      "auxiliary_loss_clip": 0.01156721,
      "auxiliary_loss_mlp": 0.01034073,
      "balance_loss_clip": 0.89376771,
      "balance_loss_mlp": 1.02676558,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.506721878765777,
      "language_loss": 0.80682135,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.82872933,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 2.8417317867279053
    },
    {
      "auxiliary_loss_clip": 0.01171155,
      "auxiliary_loss_mlp": 0.01020909,
      "balance_loss_clip": 1.01057935,
      "balance_loss_mlp": 1.01373887,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 2.26752691718704,
      "language_loss": 0.8347832,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.85670388,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 2.7215418815612793
    },
    {
      "auxiliary_loss_clip": 0.01166566,
      "auxiliary_loss_mlp": 0.01021873,
      "balance_loss_clip": 0.97219211,
      "balance_loss_mlp": 1.01429772,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.1813577587380877,
      "language_loss": 0.75630665,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.77819109,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.7013943195343018
    },
    {
      "auxiliary_loss_clip": 0.01160037,
      "auxiliary_loss_mlp": 0.01024154,
      "balance_loss_clip": 1.00845695,
      "balance_loss_mlp": 1.01771402,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.5197872155756567,
      "language_loss": 0.75230145,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.7741434,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 2.719865322113037
    },
    {
      "auxiliary_loss_clip": 0.01173227,
      "auxiliary_loss_mlp": 0.01020544,
      "balance_loss_clip": 1.0125277,
      "balance_loss_mlp": 1.01308155,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 1.5287423981354444,
      "language_loss": 0.67951965,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70145732,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 2.653977394104004
    },
    {
      "auxiliary_loss_clip": 0.01170814,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 1.01324689,
      "balance_loss_mlp": 1.02010918,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 1.9175818729574465,
      "language_loss": 0.85229999,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.87428045,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.784510850906372
    },
    {
      "auxiliary_loss_clip": 0.0117063,
      "auxiliary_loss_mlp": 0.01026546,
      "balance_loss_clip": 1.01076412,
      "balance_loss_mlp": 1.01963758,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 2.0245623226238405,
      "language_loss": 0.82278752,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84475929,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.7111363410949707
    },
    {
      "auxiliary_loss_clip": 0.01163902,
      "auxiliary_loss_mlp": 0.01024327,
      "balance_loss_clip": 0.93138313,
      "balance_loss_mlp": 1.0171628,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 1.8856742014989611,
      "language_loss": 0.83878368,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86066592,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.68186616897583
    },
    {
      "auxiliary_loss_clip": 0.01075304,
      "auxiliary_loss_mlp": 0.01001639,
      "balance_loss_clip": 0.89916396,
      "balance_loss_mlp": 0.99991095,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7134818883717962,
      "language_loss": 0.54319298,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.5639624,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.3168623447418213
    },
    {
      "auxiliary_loss_clip": 0.01166158,
      "auxiliary_loss_mlp": 0.01123273,
      "balance_loss_clip": 0.97269589,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 2.3602561117818377,
      "language_loss": 0.7823168,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.80521113,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.7495954036712646
    },
    {
      "auxiliary_loss_clip": 0.01166963,
      "auxiliary_loss_mlp": 0.0102777,
      "balance_loss_clip": 0.97248161,
      "balance_loss_mlp": 1.02039671,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 1.750084817748359,
      "language_loss": 0.70769453,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.72964185,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 3.615046977996826
    },
    {
      "auxiliary_loss_clip": 0.01166714,
      "auxiliary_loss_mlp": 0.01025664,
      "balance_loss_clip": 1.05023694,
      "balance_loss_mlp": 1.01854682,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 1.9003213676499648,
      "language_loss": 0.62898803,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.65091181,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 2.748526096343994
    },
    {
      "auxiliary_loss_clip": 0.01166481,
      "auxiliary_loss_mlp": 0.01026008,
      "balance_loss_clip": 0.85298073,
      "balance_loss_mlp": 1.0179224,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.36992648927277,
      "language_loss": 0.74051678,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76244164,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 2.803330659866333
    },
    {
      "auxiliary_loss_clip": 0.01170628,
      "auxiliary_loss_mlp": 0.01027329,
      "balance_loss_clip": 1.01069927,
      "balance_loss_mlp": 1.01985478,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 1.7650515614235138,
      "language_loss": 0.73289055,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75487012,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.661832094192505
    },
    {
      "auxiliary_loss_clip": 0.01164049,
      "auxiliary_loss_mlp": 0.01029781,
      "balance_loss_clip": 0.93339187,
      "balance_loss_mlp": 1.02234876,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 2.068368503800279,
      "language_loss": 0.6269511,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.64888936,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 4.575345277786255
    },
    {
      "auxiliary_loss_clip": 0.01162358,
      "auxiliary_loss_mlp": 0.01026232,
      "balance_loss_clip": 0.97084928,
      "balance_loss_mlp": 1.01912117,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 1.6995014280199843,
      "language_loss": 0.72105473,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.74294066,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 3.615713119506836
    },
    {
      "auxiliary_loss_clip": 0.01170261,
      "auxiliary_loss_mlp": 0.01027067,
      "balance_loss_clip": 1.04847443,
      "balance_loss_mlp": 1.01909244,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 2.6465682542545057,
      "language_loss": 0.7237041,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74567747,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.643151044845581
    },
    {
      "auxiliary_loss_clip": 0.01171731,
      "auxiliary_loss_mlp": 0.010248,
      "balance_loss_clip": 0.89419955,
      "balance_loss_mlp": 1.01700962,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.5422631297139948,
      "language_loss": 0.75588274,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.77784801,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.713472843170166
    },
    {
      "auxiliary_loss_clip": 0.01170493,
      "auxiliary_loss_mlp": 0.01026989,
      "balance_loss_clip": 1.04969382,
      "balance_loss_mlp": 1.01958609,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 3.115269906363137,
      "language_loss": 0.73624718,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75822198,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.6487700939178467
    },
    {
      "auxiliary_loss_clip": 0.01173544,
      "auxiliary_loss_mlp": 0.01024122,
      "balance_loss_clip": 0.97357553,
      "balance_loss_mlp": 1.01691556,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.2874799635141407,
      "language_loss": 0.71286714,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.73484385,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.593489408493042
    },
    {
      "auxiliary_loss_clip": 0.01169221,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 0.93255359,
      "balance_loss_mlp": 1.0223906,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 1.7005645013740507,
      "language_loss": 0.78346884,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80546117,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.7173421382904053
    },
    {
      "auxiliary_loss_clip": 0.01160118,
      "auxiliary_loss_mlp": 0.01031061,
      "balance_loss_clip": 1.00864506,
      "balance_loss_mlp": 1.02374744,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.4903908026683685,
      "language_loss": 0.81969571,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.84160757,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 2.679100513458252
    },
    {
      "auxiliary_loss_clip": 0.01064471,
      "auxiliary_loss_mlp": 0.0100319,
      "balance_loss_clip": 1.01142502,
      "balance_loss_mlp": 1.00150955,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7674527856937771,
      "language_loss": 0.56927282,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58994943,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.2694151401519775
    },
    {
      "auxiliary_loss_clip": 0.01150708,
      "auxiliary_loss_mlp": 0.01025255,
      "balance_loss_clip": 0.89402497,
      "balance_loss_mlp": 1.01785851,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 1.7307242685740476,
      "language_loss": 0.71501195,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73677164,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 2.788404703140259
    },
    {
      "auxiliary_loss_clip": 0.01161346,
      "auxiliary_loss_mlp": 0.01024067,
      "balance_loss_clip": 0.97177684,
      "balance_loss_mlp": 1.01768637,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.7258040522047131,
      "language_loss": 0.63797563,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.65982974,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 2.7616562843322754
    },
    {
      "auxiliary_loss_clip": 0.01165188,
      "auxiliary_loss_mlp": 0.01024918,
      "balance_loss_clip": 0.96771103,
      "balance_loss_mlp": 1.01793206,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 1.8577517841191349,
      "language_loss": 0.68982649,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.71172762,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.731534957885742
    },
    {
      "auxiliary_loss_clip": 0.01171678,
      "auxiliary_loss_mlp": 0.01026617,
      "balance_loss_clip": 1.05057716,
      "balance_loss_mlp": 1.01935995,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 1.9054867289327506,
      "language_loss": 0.75520051,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.77718353,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 2.648725986480713
    },
    {
      "auxiliary_loss_clip": 0.01165206,
      "auxiliary_loss_mlp": 0.01031672,
      "balance_loss_clip": 0.93458378,
      "balance_loss_mlp": 1.02400696,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 2.158352637637883,
      "language_loss": 0.77439654,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.79636526,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.74403977394104
    },
    {
      "auxiliary_loss_clip": 0.01161819,
      "auxiliary_loss_mlp": 0.010259,
      "balance_loss_clip": 0.97034121,
      "balance_loss_mlp": 1.01818991,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 2.3491469915688494,
      "language_loss": 0.82899415,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85087132,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 2.694885015487671
    },
    {
      "auxiliary_loss_clip": 0.01170835,
      "auxiliary_loss_mlp": 0.01023663,
      "balance_loss_clip": 1.0503006,
      "balance_loss_mlp": 1.01664138,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 1.6191863174410788,
      "language_loss": 0.68897313,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.71091813,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.6705281734466553
    },
    {
      "auxiliary_loss_clip": 0.01170875,
      "auxiliary_loss_mlp": 0.01022898,
      "balance_loss_clip": 1.00702214,
      "balance_loss_mlp": 1.01548934,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 1.9573976680866985,
      "language_loss": 0.77341789,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79535556,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 2.6678662300109863
    },
    {
      "auxiliary_loss_clip": 0.01161199,
      "auxiliary_loss_mlp": 0.01025835,
      "balance_loss_clip": 0.93327028,
      "balance_loss_mlp": 1.01765108,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 5.441016320482679,
      "language_loss": 0.65402806,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.67589837,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.699707269668579
    },
    {
      "auxiliary_loss_clip": 0.01167604,
      "auxiliary_loss_mlp": 0.01022888,
      "balance_loss_clip": 0.97116297,
      "balance_loss_mlp": 1.01629901,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 1.5894874863636954,
      "language_loss": 0.77104801,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.7929529,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 2.7509653568267822
    },
    {
      "auxiliary_loss_clip": 0.01161576,
      "auxiliary_loss_mlp": 0.01122447,
      "balance_loss_clip": 0.9710685,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 1.7541450220167043,
      "language_loss": 0.79392421,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81676435,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.731595993041992
    },
    {
      "auxiliary_loss_clip": 0.01164118,
      "auxiliary_loss_mlp": 0.01022662,
      "balance_loss_clip": 1.0074439,
      "balance_loss_mlp": 1.01579845,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 2.729360460139407,
      "language_loss": 0.62436879,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.64623654,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 2.751990795135498
    },
    {
      "auxiliary_loss_clip": 0.0115774,
      "auxiliary_loss_mlp": 0.01025766,
      "balance_loss_clip": 0.93396223,
      "balance_loss_mlp": 1.01825595,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.375604346058042,
      "language_loss": 0.67778313,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.69961816,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.7335317134857178
    },
    {
      "auxiliary_loss_clip": 0.01160288,
      "auxiliary_loss_mlp": 0.01028313,
      "balance_loss_clip": 0.9723897,
      "balance_loss_mlp": 1.02174461,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.8386052092834513,
      "language_loss": 0.77489895,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.796785,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 3.7282068729400635
    },
    {
      "auxiliary_loss_clip": 0.01168227,
      "auxiliary_loss_mlp": 0.01030424,
      "balance_loss_clip": 1.01079524,
      "balance_loss_mlp": 1.02314377,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 2.7380550344136587,
      "language_loss": 0.77744782,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.7994343,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.6499505043029785
    },
    {
      "auxiliary_loss_clip": 0.01160916,
      "auxiliary_loss_mlp": 0.01024478,
      "balance_loss_clip": 1.00717735,
      "balance_loss_mlp": 1.01714313,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 2.967518930171413,
      "language_loss": 0.4844487,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.50630265,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.722121238708496
    },
    {
      "auxiliary_loss_clip": 0.01168932,
      "auxiliary_loss_mlp": 0.0102662,
      "balance_loss_clip": 1.04773045,
      "balance_loss_mlp": 1.01939344,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 2.615681143278034,
      "language_loss": 0.82718396,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.84913945,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 3.6207034587860107
    },
    {
      "auxiliary_loss_clip": 0.01063992,
      "auxiliary_loss_mlp": 0.01002336,
      "balance_loss_clip": 1.01107001,
      "balance_loss_mlp": 1.00071454,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.7579113342173114,
      "language_loss": 0.55088472,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57154799,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 3.9474401473999023
    },
    {
      "auxiliary_loss_clip": 0.01168287,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 0.89485931,
      "balance_loss_mlp": 1.02077281,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 1.9604808810351924,
      "language_loss": 0.68114436,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.7031073,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 3.6220667362213135
    },
    {
      "auxiliary_loss_clip": 0.01165064,
      "auxiliary_loss_mlp": 0.01021572,
      "balance_loss_clip": 0.97035265,
      "balance_loss_mlp": 1.01474094,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 1.9222076596784017,
      "language_loss": 0.70890605,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.73077244,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.7332510948181152
    },
    {
      "auxiliary_loss_clip": 0.01172177,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 0.97124046,
      "balance_loss_mlp": 1.01925099,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 2.224996591295898,
      "language_loss": 0.74996555,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77195513,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.782536506652832
    },
    {
      "auxiliary_loss_clip": 0.01067289,
      "auxiliary_loss_mlp": 0.0100211,
      "balance_loss_clip": 0.973508,
      "balance_loss_mlp": 1.00041699,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.8178059965715253,
      "language_loss": 0.59364504,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61433899,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 3.075922966003418
    },
    {
      "auxiliary_loss_clip": 0.01167205,
      "auxiliary_loss_mlp": 0.01023005,
      "balance_loss_clip": 1.00996256,
      "balance_loss_mlp": 1.01544738,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 1.9171738794588047,
      "language_loss": 0.80183601,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.8237381,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.6321747303009033
    },
    {
      "auxiliary_loss_clip": 0.01168232,
      "auxiliary_loss_mlp": 0.01032765,
      "balance_loss_clip": 1.01094186,
      "balance_loss_mlp": 1.02521324,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 12.33746865964735,
      "language_loss": 0.77087903,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.792889,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.6050546169281006
    },
    {
      "auxiliary_loss_clip": 0.0116545,
      "auxiliary_loss_mlp": 0.01027642,
      "balance_loss_clip": 0.96901375,
      "balance_loss_mlp": 1.02057242,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 2.4933195995171693,
      "language_loss": 0.92384493,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94577587,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 2.6627719402313232
    },
    {
      "auxiliary_loss_clip": 0.01154336,
      "auxiliary_loss_mlp": 0.01030017,
      "balance_loss_clip": 0.9313767,
      "balance_loss_mlp": 1.02226257,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.0267394482219334,
      "language_loss": 0.87286311,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.89470667,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 2.7594470977783203
    },
    {
      "auxiliary_loss_clip": 0.01152696,
      "auxiliary_loss_mlp": 0.01022862,
      "balance_loss_clip": 0.92934668,
      "balance_loss_mlp": 1.01607013,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.508026135950106,
      "language_loss": 0.79777324,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.81952882,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 2.681849479675293
    },
    {
      "auxiliary_loss_clip": 0.01175096,
      "auxiliary_loss_mlp": 0.01025133,
      "balance_loss_clip": 0.89475691,
      "balance_loss_mlp": 1.01818967,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 1.6868219143426835,
      "language_loss": 0.77604818,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79805052,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 2.737743377685547
    },
    {
      "auxiliary_loss_clip": 0.0116604,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 0.97452676,
      "balance_loss_mlp": 1.01844001,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 1.7992371205862039,
      "language_loss": 0.76866382,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.79057878,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.65386962890625
    },
    {
      "auxiliary_loss_clip": 0.01150166,
      "auxiliary_loss_mlp": 0.01026342,
      "balance_loss_clip": 0.96778989,
      "balance_loss_mlp": 1.0192852,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 2.0901184783078373,
      "language_loss": 0.75892651,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.78069162,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 2.7222819328308105
    },
    {
      "auxiliary_loss_clip": 0.01160616,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 0.93176353,
      "balance_loss_mlp": 1.01934958,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.512944503423628,
      "language_loss": 0.78188258,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.80375648,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 2.7240378856658936
    },
    {
      "auxiliary_loss_clip": 0.0116969,
      "auxiliary_loss_mlp": 0.01022069,
      "balance_loss_clip": 0.93466645,
      "balance_loss_mlp": 1.01470172,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 1.932607883357991,
      "language_loss": 0.7937845,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.81570208,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.6483469009399414
    },
    {
      "auxiliary_loss_clip": 0.01163871,
      "auxiliary_loss_mlp": 0.0102279,
      "balance_loss_clip": 0.89604676,
      "balance_loss_mlp": 1.01567364,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 2.0726915182496066,
      "language_loss": 0.74918747,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.77105409,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.7471606731414795
    },
    {
      "auxiliary_loss_clip": 0.01165388,
      "auxiliary_loss_mlp": 0.01031233,
      "balance_loss_clip": 0.97498369,
      "balance_loss_mlp": 1.02363932,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 1.8523360567322353,
      "language_loss": 0.81833923,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.84030545,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 2.6175625324249268
    },
    {
      "auxiliary_loss_clip": 0.01166159,
      "auxiliary_loss_mlp": 0.01023199,
      "balance_loss_clip": 1.0110482,
      "balance_loss_mlp": 1.01661873,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 2.0099038253897525,
      "language_loss": 0.88931823,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.91121179,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.6505024433135986
    },
    {
      "auxiliary_loss_clip": 0.01067485,
      "auxiliary_loss_mlp": 0.01001045,
      "balance_loss_clip": 0.97430515,
      "balance_loss_mlp": 0.9994716,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 1.1429555503756774,
      "language_loss": 0.60759544,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62828076,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 3.086188554763794
    },
    {
      "auxiliary_loss_clip": 0.01162682,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 0.92932862,
      "balance_loss_mlp": 1.02278161,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 2.7773817193574954,
      "language_loss": 0.78744984,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.809385,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 2.7159907817840576
    },
    {
      "auxiliary_loss_clip": 0.01168945,
      "auxiliary_loss_mlp": 0.0102354,
      "balance_loss_clip": 1.01085103,
      "balance_loss_mlp": 1.01659584,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.05838174990164,
      "language_loss": 0.64894676,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.67087162,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 2.72501802444458
    },
    {
      "auxiliary_loss_clip": 0.01066717,
      "auxiliary_loss_mlp": 0.01002022,
      "balance_loss_clip": 0.97403347,
      "balance_loss_mlp": 1.00037718,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7470037645214519,
      "language_loss": 0.51844645,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53913391,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.105548620223999
    },
    {
      "auxiliary_loss_clip": 0.01159884,
      "auxiliary_loss_mlp": 0.01023424,
      "balance_loss_clip": 0.96989691,
      "balance_loss_mlp": 1.01618791,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 1.9189862839858003,
      "language_loss": 0.68786263,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.7096957,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 3.699991226196289
    },
    {
      "auxiliary_loss_clip": 0.01174877,
      "auxiliary_loss_mlp": 0.0102481,
      "balance_loss_clip": 0.97563142,
      "balance_loss_mlp": 1.01711774,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 2.0294466539336717,
      "language_loss": 0.74324226,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76523912,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.663874864578247
    },
    {
      "auxiliary_loss_clip": 0.0117397,
      "auxiliary_loss_mlp": 0.01028065,
      "balance_loss_clip": 0.9751876,
      "balance_loss_mlp": 1.02068281,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.6439204689366957,
      "language_loss": 0.85042328,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87244362,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.66341233253479
    },
    {
      "auxiliary_loss_clip": 0.01169446,
      "auxiliary_loss_mlp": 0.01122818,
      "balance_loss_clip": 0.89663142,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 2.751365446721865,
      "language_loss": 0.70482469,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.72774726,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 3.831052541732788
    },
    {
      "auxiliary_loss_clip": 0.01169777,
      "auxiliary_loss_mlp": 0.01024753,
      "balance_loss_clip": 1.01108789,
      "balance_loss_mlp": 1.01773763,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 1.7790812195834995,
      "language_loss": 0.71632826,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.73827362,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 3.595609188079834
    },
    {
      "auxiliary_loss_clip": 0.01148436,
      "auxiliary_loss_mlp": 0.01024265,
      "balance_loss_clip": 0.81562257,
      "balance_loss_mlp": 1.01735711,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 2.0405494389312655,
      "language_loss": 0.83836722,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.86009425,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.936506748199463
    },
    {
      "auxiliary_loss_clip": 0.01157333,
      "auxiliary_loss_mlp": 0.01024995,
      "balance_loss_clip": 0.93151039,
      "balance_loss_mlp": 1.01729417,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 1.8425974359463073,
      "language_loss": 0.78273606,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80455935,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 3.9169740676879883
    },
    {
      "auxiliary_loss_clip": 0.01173087,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.04932404,
      "balance_loss_mlp": 1.01980913,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 2.0390480189828306,
      "language_loss": 0.7426911,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76469582,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.5745086669921875
    },
    {
      "auxiliary_loss_clip": 0.011569,
      "auxiliary_loss_mlp": 0.01020546,
      "balance_loss_clip": 1.00702655,
      "balance_loss_mlp": 1.013906,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.700240856593312,
      "language_loss": 0.77773726,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.79951167,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.664926528930664
    },
    {
      "auxiliary_loss_clip": 0.01167642,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.0497632,
      "balance_loss_mlp": 1.01997495,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 2.009101331150661,
      "language_loss": 0.79998517,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82192814,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.624457359313965
    },
    {
      "auxiliary_loss_clip": 0.01170322,
      "auxiliary_loss_mlp": 0.01021704,
      "balance_loss_clip": 0.97044498,
      "balance_loss_mlp": 1.01489162,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.993130302174788,
      "language_loss": 0.70075834,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.72267854,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.6223392486572266
    },
    {
      "auxiliary_loss_clip": 0.01176844,
      "auxiliary_loss_mlp": 0.01031456,
      "balance_loss_clip": 0.93331563,
      "balance_loss_mlp": 1.02310538,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 3.3362264907761174,
      "language_loss": 0.67142278,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.69350576,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 2.675055742263794
    },
    {
      "auxiliary_loss_clip": 0.01168943,
      "auxiliary_loss_mlp": 0.01122771,
      "balance_loss_clip": 1.04860306,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 2.2102934009845576,
      "language_loss": 0.62744182,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.65035897,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 2.638899803161621
    },
    {
      "auxiliary_loss_clip": 0.01169925,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.04981208,
      "balance_loss_mlp": 1.01966977,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 2.2526476574369494,
      "language_loss": 0.78784657,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.8098141,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.579324960708618
    },
    {
      "auxiliary_loss_clip": 0.01162491,
      "auxiliary_loss_mlp": 0.01024437,
      "balance_loss_clip": 1.00694966,
      "balance_loss_mlp": 1.01750183,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.6510658916158942,
      "language_loss": 0.69174808,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71361738,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 2.60844349861145
    },
    {
      "auxiliary_loss_clip": 0.0117087,
      "auxiliary_loss_mlp": 0.01031018,
      "balance_loss_clip": 0.97622246,
      "balance_loss_mlp": 1.02273929,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 1.8563142945198647,
      "language_loss": 0.70203584,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72405469,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.684690475463867
    },
    {
      "auxiliary_loss_clip": 0.01173528,
      "auxiliary_loss_mlp": 0.01025846,
      "balance_loss_clip": 1.00985897,
      "balance_loss_mlp": 1.01776958,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 1.756533577385092,
      "language_loss": 0.82491338,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84690714,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 2.6555182933807373
    },
    {
      "auxiliary_loss_clip": 0.01170047,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 0.97328126,
      "balance_loss_mlp": 1.01952434,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 39.89607507396739,
      "language_loss": 0.70682108,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.72879219,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.7861523628234863
    },
    {
      "auxiliary_loss_clip": 0.01166268,
      "auxiliary_loss_mlp": 0.01022738,
      "balance_loss_clip": 0.97083384,
      "balance_loss_mlp": 1.01535296,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 2.884961164328366,
      "language_loss": 0.61503285,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.6369229,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 2.633899211883545
    },
    {
      "auxiliary_loss_clip": 0.01067349,
      "auxiliary_loss_mlp": 0.01000716,
      "balance_loss_clip": 0.90104449,
      "balance_loss_mlp": 0.99914211,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8490035114805109,
      "language_loss": 0.6258328,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64651346,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.3796374797821045
    },
    {
      "auxiliary_loss_clip": 0.01168012,
      "auxiliary_loss_mlp": 0.01027394,
      "balance_loss_clip": 1.04770637,
      "balance_loss_mlp": 1.02057195,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 1.8421484660248946,
      "language_loss": 0.74267566,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76462972,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 2.8123090267181396
    },
    {
      "auxiliary_loss_clip": 0.0116167,
      "auxiliary_loss_mlp": 0.01026352,
      "balance_loss_clip": 0.93062413,
      "balance_loss_mlp": 1.01906228,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 1.9898719350929364,
      "language_loss": 0.65924209,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.6811223,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.717527389526367
    },
    {
      "auxiliary_loss_clip": 0.01161098,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 0.97071433,
      "balance_loss_mlp": 1.02317178,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 7.569128756046113,
      "language_loss": 0.56654215,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.58846247,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.6879334449768066
    },
    {
      "auxiliary_loss_clip": 0.01175534,
      "auxiliary_loss_mlp": 0.01024852,
      "balance_loss_clip": 0.93405116,
      "balance_loss_mlp": 1.01762235,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 2.086039644443325,
      "language_loss": 0.79194432,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.81394815,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 2.713975667953491
    },
    {
      "auxiliary_loss_clip": 0.01167787,
      "auxiliary_loss_mlp": 0.010232,
      "balance_loss_clip": 0.9353326,
      "balance_loss_mlp": 1.01607728,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.6432225615358553,
      "language_loss": 0.76122284,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.78313273,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.73264479637146
    },
    {
      "auxiliary_loss_clip": 0.01166812,
      "auxiliary_loss_mlp": 0.01027431,
      "balance_loss_clip": 1.00954342,
      "balance_loss_mlp": 1.02092183,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 1.6126657808148752,
      "language_loss": 0.77318394,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79512638,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.613672971725464
    },
    {
      "auxiliary_loss_clip": 0.01060711,
      "auxiliary_loss_mlp": 0.01003402,
      "balance_loss_clip": 0.89924037,
      "balance_loss_mlp": 1.00166178,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6578113169706394,
      "language_loss": 0.58036065,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.6010018,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 4.272123098373413
    },
    {
      "auxiliary_loss_clip": 0.01170096,
      "auxiliary_loss_mlp": 0.01041215,
      "balance_loss_clip": 0.9729563,
      "balance_loss_mlp": 1.03347182,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 2.082836240916339,
      "language_loss": 0.63554978,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65766287,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.7535018920898438
    },
    {
      "auxiliary_loss_clip": 0.01170149,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.04842949,
      "balance_loss_mlp": 1.01944613,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 1.991350267421015,
      "language_loss": 0.77155054,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79352558,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.63454532623291
    },
    {
      "auxiliary_loss_clip": 0.01171197,
      "auxiliary_loss_mlp": 0.01021634,
      "balance_loss_clip": 1.04991293,
      "balance_loss_mlp": 1.01445723,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 3.139264498811876,
      "language_loss": 0.81561929,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.83754754,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.692244529724121
    },
    {
      "auxiliary_loss_clip": 0.0117125,
      "auxiliary_loss_mlp": 0.01031435,
      "balance_loss_clip": 1.04986405,
      "balance_loss_mlp": 1.02363849,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 2.007149800148861,
      "language_loss": 0.74365646,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76568329,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 4.504699945449829
    },
    {
      "auxiliary_loss_clip": 0.01167866,
      "auxiliary_loss_mlp": 0.01025439,
      "balance_loss_clip": 0.97231948,
      "balance_loss_mlp": 1.01860845,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 1.7932265210962153,
      "language_loss": 0.74735171,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.76928478,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.7881267070770264
    },
    {
      "auxiliary_loss_clip": 0.0114982,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 0.85518634,
      "balance_loss_mlp": 1.02125907,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 3.290810310578613,
      "language_loss": 0.64276814,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.66455221,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 3.8020474910736084
    },
    {
      "auxiliary_loss_clip": 0.01170335,
      "auxiliary_loss_mlp": 0.01026978,
      "balance_loss_clip": 1.00993848,
      "balance_loss_mlp": 1.01902676,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.2478257302069315,
      "language_loss": 0.78599113,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.80796432,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.7174370288848877
    },
    {
      "auxiliary_loss_clip": 0.01169293,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 1.04795432,
      "balance_loss_mlp": 1.01987123,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 1.9371584873492362,
      "language_loss": 0.73863184,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.76059198,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.630156993865967
    },
    {
      "auxiliary_loss_clip": 0.01166171,
      "auxiliary_loss_mlp": 0.01020966,
      "balance_loss_clip": 1.04941022,
      "balance_loss_mlp": 1.01418042,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 2.548764480790678,
      "language_loss": 0.7553407,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.77721208,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.6325976848602295
    },
    {
      "auxiliary_loss_clip": 0.01168157,
      "auxiliary_loss_mlp": 0.0112205,
      "balance_loss_clip": 0.93393254,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.6576658391345664,
      "language_loss": 0.74099928,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.76390135,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.7096660137176514
    },
    {
      "auxiliary_loss_clip": 0.01160175,
      "auxiliary_loss_mlp": 0.0102582,
      "balance_loss_clip": 0.97073054,
      "balance_loss_mlp": 1.01881373,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 1.966428204889125,
      "language_loss": 0.81088638,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.83274633,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 2.7209115028381348
    },
    {
      "auxiliary_loss_clip": 0.011676,
      "auxiliary_loss_mlp": 0.01030081,
      "balance_loss_clip": 1.00917327,
      "balance_loss_mlp": 1.02270794,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.1111548981822916,
      "language_loss": 0.72504413,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.7470209,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 2.5723042488098145
    },
    {
      "auxiliary_loss_clip": 0.01167873,
      "auxiliary_loss_mlp": 0.01024159,
      "balance_loss_clip": 0.97667623,
      "balance_loss_mlp": 1.01729226,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.7941830223224111,
      "language_loss": 0.83633274,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.858253,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.6353025436401367
    },
    {
      "auxiliary_loss_clip": 0.0116717,
      "auxiliary_loss_mlp": 0.01029425,
      "balance_loss_clip": 0.93186831,
      "balance_loss_mlp": 1.02223325,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 1.4220665977843678,
      "language_loss": 0.67397034,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69593626,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 2.7359561920166016
    },
    {
      "auxiliary_loss_clip": 0.01174404,
      "auxiliary_loss_mlp": 0.01121719,
      "balance_loss_clip": 0.93490326,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 1.6532059043279839,
      "language_loss": 0.83190441,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.85486567,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.770930290222168
    },
    {
      "auxiliary_loss_clip": 0.01170527,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 0.93586296,
      "balance_loss_mlp": 1.01865673,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.2154646618305036,
      "language_loss": 0.62643582,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.64840198,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 2.73699688911438
    },
    {
      "auxiliary_loss_clip": 0.0116801,
      "auxiliary_loss_mlp": 0.01025615,
      "balance_loss_clip": 0.89996493,
      "balance_loss_mlp": 1.01797926,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 2.1340543469765434,
      "language_loss": 0.7526083,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77454454,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 2.7565829753875732
    },
    {
      "auxiliary_loss_clip": 0.01166469,
      "auxiliary_loss_mlp": 0.01021275,
      "balance_loss_clip": 0.97304165,
      "balance_loss_mlp": 1.01394975,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.5111778652724324,
      "language_loss": 0.85325956,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.87513703,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 2.6684770584106445
    },
    {
      "auxiliary_loss_clip": 0.01163371,
      "auxiliary_loss_mlp": 0.01026174,
      "balance_loss_clip": 0.93111396,
      "balance_loss_mlp": 1.01883376,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 2.2800610078324746,
      "language_loss": 0.7321682,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75406361,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 2.723339796066284
    },
    {
      "auxiliary_loss_clip": 0.01161069,
      "auxiliary_loss_mlp": 0.01023366,
      "balance_loss_clip": 0.97041094,
      "balance_loss_mlp": 1.01676524,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 1.6629028460841695,
      "language_loss": 0.80590099,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.82774538,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 2.699894905090332
    },
    {
      "auxiliary_loss_clip": 0.01157939,
      "auxiliary_loss_mlp": 0.01027247,
      "balance_loss_clip": 0.89483762,
      "balance_loss_mlp": 1.02052665,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 1.8688957462196218,
      "language_loss": 0.75357831,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.7754302,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.839139461517334
    },
    {
      "auxiliary_loss_clip": 0.01167163,
      "auxiliary_loss_mlp": 0.01122032,
      "balance_loss_clip": 1.00887346,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 1.7464122530970434,
      "language_loss": 0.83195508,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.85484707,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.758955955505371
    },
    {
      "auxiliary_loss_clip": 0.01171289,
      "auxiliary_loss_mlp": 0.01024936,
      "balance_loss_clip": 0.97165871,
      "balance_loss_mlp": 1.01763141,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 2.3021402742701182,
      "language_loss": 0.76061934,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78258163,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.709059000015259
    },
    {
      "auxiliary_loss_clip": 0.01171062,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.01025033,
      "balance_loss_mlp": 1.01820695,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 2.3791925065779878,
      "language_loss": 0.71758258,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.73954785,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 2.6499342918395996
    },
    {
      "auxiliary_loss_clip": 0.01167151,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 0.93526292,
      "balance_loss_mlp": 1.01889563,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 1.7587512685523554,
      "language_loss": 0.7873919,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.80932164,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 3.6068596839904785
    },
    {
      "auxiliary_loss_clip": 0.01138624,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 0.85515398,
      "balance_loss_mlp": 1.01756215,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 1.6772272976519391,
      "language_loss": 0.76199389,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78362262,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.780548334121704
    },
    {
      "auxiliary_loss_clip": 0.01165073,
      "auxiliary_loss_mlp": 0.01026268,
      "balance_loss_clip": 0.93262494,
      "balance_loss_mlp": 1.01890647,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 1.6677657578171343,
      "language_loss": 0.81851375,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.84042716,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.8313333988189697
    },
    {
      "auxiliary_loss_clip": 0.01167085,
      "auxiliary_loss_mlp": 0.01023486,
      "balance_loss_clip": 0.9309361,
      "balance_loss_mlp": 1.01641655,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 2.035965235422022,
      "language_loss": 0.85563135,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.87753713,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.677197217941284
    },
    {
      "auxiliary_loss_clip": 0.01168108,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.0110662,
      "balance_loss_mlp": 1.01757574,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 1.8470907327375574,
      "language_loss": 0.7953527,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81728721,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.6642167568206787
    },
    {
      "auxiliary_loss_clip": 0.01164962,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 0.89559525,
      "balance_loss_mlp": 1.02049398,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 1.6291558912236208,
      "language_loss": 0.78178793,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80370754,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 3.637587547302246
    },
    {
      "auxiliary_loss_clip": 0.01162218,
      "auxiliary_loss_mlp": 0.01026602,
      "balance_loss_clip": 1.01034749,
      "balance_loss_mlp": 1.01890373,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 2.107885679957906,
      "language_loss": 0.80068672,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82257491,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.641528367996216
    },
    {
      "auxiliary_loss_clip": 0.01174313,
      "auxiliary_loss_mlp": 0.01027158,
      "balance_loss_clip": 1.01455951,
      "balance_loss_mlp": 1.02027655,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 1.7222961882854737,
      "language_loss": 0.56920528,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.59122002,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 3.65535306930542
    },
    {
      "auxiliary_loss_clip": 0.01169042,
      "auxiliary_loss_mlp": 0.01026946,
      "balance_loss_clip": 0.97372508,
      "balance_loss_mlp": 1.01886976,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 2.554209653008748,
      "language_loss": 0.67029464,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69225454,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.6538498401641846
    },
    {
      "auxiliary_loss_clip": 0.01169054,
      "auxiliary_loss_mlp": 0.01027063,
      "balance_loss_clip": 1.04915786,
      "balance_loss_mlp": 1.01976442,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 1.640587555740948,
      "language_loss": 0.68846458,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.71042573,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.678819417953491
    },
    {
      "auxiliary_loss_clip": 0.01169407,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.0102644,
      "balance_loss_mlp": 1.02089214,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.769448411754872,
      "language_loss": 0.72201788,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74399042,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.6846542358398438
    },
    {
      "auxiliary_loss_clip": 0.01171574,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 0.97114253,
      "balance_loss_mlp": 1.02249658,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 4.513796967763238,
      "language_loss": 0.82332379,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84533954,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.66152024269104
    },
    {
      "auxiliary_loss_clip": 0.01171058,
      "auxiliary_loss_mlp": 0.01023221,
      "balance_loss_clip": 1.01100397,
      "balance_loss_mlp": 1.01622987,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.5857881770893578,
      "language_loss": 0.70374483,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72568762,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 2.6479439735412598
    },
    {
      "auxiliary_loss_clip": 0.01162624,
      "auxiliary_loss_mlp": 0.01023998,
      "balance_loss_clip": 0.9331997,
      "balance_loss_mlp": 1.01628864,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.7916918894339633,
      "language_loss": 0.81966704,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.8415333,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 2.8443591594696045
    },
    {
      "auxiliary_loss_clip": 0.01170954,
      "auxiliary_loss_mlp": 0.01024864,
      "balance_loss_clip": 1.01177728,
      "balance_loss_mlp": 1.01833427,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 3.6730258778756864,
      "language_loss": 0.68626463,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.70822287,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 2.675081968307495
    },
    {
      "auxiliary_loss_clip": 0.01172429,
      "auxiliary_loss_mlp": 0.01024997,
      "balance_loss_clip": 1.0496726,
      "balance_loss_mlp": 1.01768398,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 1.9472473253774718,
      "language_loss": 0.61737907,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.63935333,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 2.581887722015381
    },
    {
      "auxiliary_loss_clip": 0.01163821,
      "auxiliary_loss_mlp": 0.0102736,
      "balance_loss_clip": 0.93260086,
      "balance_loss_mlp": 1.02050257,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 3.5170414591784906,
      "language_loss": 0.75320458,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.77511632,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.697286605834961
    },
    {
      "auxiliary_loss_clip": 0.0117549,
      "auxiliary_loss_mlp": 0.01025932,
      "balance_loss_clip": 0.89682627,
      "balance_loss_mlp": 1.01900899,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 1.5341427643215397,
      "language_loss": 0.72894323,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.75095737,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 2.7489545345306396
    },
    {
      "auxiliary_loss_clip": 0.01171126,
      "auxiliary_loss_mlp": 0.01028986,
      "balance_loss_clip": 1.00930476,
      "balance_loss_mlp": 1.02179742,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 1.580088798784451,
      "language_loss": 0.79930085,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.82130194,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.633277177810669
    },
    {
      "auxiliary_loss_clip": 0.01063292,
      "auxiliary_loss_mlp": 0.01003662,
      "balance_loss_clip": 1.01038051,
      "balance_loss_mlp": 1.0020411,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.862206761206234,
      "language_loss": 0.60204744,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.62271702,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.197417974472046
    },
    {
      "auxiliary_loss_clip": 0.01163411,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 0.97007096,
      "balance_loss_mlp": 1.01827788,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.6388433094675248,
      "language_loss": 0.7558794,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.77776664,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.717078685760498
    },
    {
      "auxiliary_loss_clip": 0.01172081,
      "auxiliary_loss_mlp": 0.01026993,
      "balance_loss_clip": 0.93291378,
      "balance_loss_mlp": 1.01906562,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 1.754695751155953,
      "language_loss": 0.68672049,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.70871127,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 2.713456153869629
    },
    {
      "auxiliary_loss_clip": 0.01169477,
      "auxiliary_loss_mlp": 0.01021369,
      "balance_loss_clip": 1.04704499,
      "balance_loss_mlp": 1.013901,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 2.221681840857095,
      "language_loss": 0.79631877,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.81822717,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 2.5815236568450928
    },
    {
      "auxiliary_loss_clip": 0.01168605,
      "auxiliary_loss_mlp": 0.0102878,
      "balance_loss_clip": 1.04833138,
      "balance_loss_mlp": 1.02214646,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 2.1499872659105925,
      "language_loss": 0.88035035,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90232426,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.569432497024536
    },
    {
      "auxiliary_loss_clip": 0.0116703,
      "auxiliary_loss_mlp": 0.0102222,
      "balance_loss_clip": 0.97033763,
      "balance_loss_mlp": 1.0148344,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 1.604988166701141,
      "language_loss": 0.70315784,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72505039,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.7270846366882324
    },
    {
      "auxiliary_loss_clip": 0.01159263,
      "auxiliary_loss_mlp": 0.01023583,
      "balance_loss_clip": 0.93094969,
      "balance_loss_mlp": 1.01652324,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 2.5889183588125593,
      "language_loss": 0.60822302,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.63005149,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.8824820518493652
    },
    {
      "auxiliary_loss_clip": 0.01162374,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 0.97173977,
      "balance_loss_mlp": 1.02017379,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 1.7502654147763814,
      "language_loss": 0.71807051,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.73996913,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 3.5997045040130615
    },
    {
      "auxiliary_loss_clip": 0.01155168,
      "auxiliary_loss_mlp": 0.01029498,
      "balance_loss_clip": 0.81239796,
      "balance_loss_mlp": 1.02257514,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 1.5936689610335135,
      "language_loss": 0.85662019,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.87846684,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 3.0266430377960205
    },
    {
      "auxiliary_loss_clip": 0.01169034,
      "auxiliary_loss_mlp": 0.01027439,
      "balance_loss_clip": 0.97082007,
      "balance_loss_mlp": 1.01997614,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 1.8051199020790916,
      "language_loss": 0.71424925,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73621398,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 3.058976888656616
    },
    {
      "auxiliary_loss_clip": 0.0116158,
      "auxiliary_loss_mlp": 0.01027629,
      "balance_loss_clip": 1.00868464,
      "balance_loss_mlp": 1.02088153,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 1.8801537510506532,
      "language_loss": 0.81839085,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.84028298,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.666015863418579
    },
    {
      "auxiliary_loss_clip": 0.01169653,
      "auxiliary_loss_mlp": 0.01024296,
      "balance_loss_clip": 0.93287343,
      "balance_loss_mlp": 1.01732194,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 2.2035345742630494,
      "language_loss": 0.77846062,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.80040008,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 3.9749510288238525
    },
    {
      "auxiliary_loss_clip": 0.01172862,
      "auxiliary_loss_mlp": 0.01032955,
      "balance_loss_clip": 0.93448889,
      "balance_loss_mlp": 1.02489972,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 1.5698079028688885,
      "language_loss": 0.65159547,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67365366,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 3.581599235534668
    },
    {
      "auxiliary_loss_clip": 0.01166755,
      "auxiliary_loss_mlp": 0.01025177,
      "balance_loss_clip": 1.04872704,
      "balance_loss_mlp": 1.01824474,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 1.7398388431109248,
      "language_loss": 0.69398546,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.71590477,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.612736940383911
    },
    {
      "auxiliary_loss_clip": 0.0117136,
      "auxiliary_loss_mlp": 0.01023722,
      "balance_loss_clip": 0.97436821,
      "balance_loss_mlp": 1.01622915,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 2.519392867465125,
      "language_loss": 0.76206094,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.78401172,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 3.6062991619110107
    },
    {
      "auxiliary_loss_clip": 0.01168509,
      "auxiliary_loss_mlp": 0.01027449,
      "balance_loss_clip": 0.97123021,
      "balance_loss_mlp": 1.02011216,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 2.8364509001636207,
      "language_loss": 0.61416662,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.63612628,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.652986526489258
    },
    {
      "auxiliary_loss_clip": 0.01171112,
      "auxiliary_loss_mlp": 0.01022554,
      "balance_loss_clip": 0.97310889,
      "balance_loss_mlp": 1.0158242,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.766024051407861,
      "language_loss": 0.89577323,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.91770983,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.7179834842681885
    },
    {
      "auxiliary_loss_clip": 0.01167358,
      "auxiliary_loss_mlp": 0.01024061,
      "balance_loss_clip": 1.00760388,
      "balance_loss_mlp": 1.01742125,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 1.6598962564248552,
      "language_loss": 0.71982104,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.74173522,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.620810031890869
    },
    {
      "auxiliary_loss_clip": 0.01170355,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 1.04984784,
      "balance_loss_mlp": 1.02014542,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 1.7259327623831213,
      "language_loss": 0.86216772,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88414228,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.6278812885284424
    },
    {
      "auxiliary_loss_clip": 0.01157738,
      "auxiliary_loss_mlp": 0.01029126,
      "balance_loss_clip": 0.96734583,
      "balance_loss_mlp": 1.02217913,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 3.5639191518689723,
      "language_loss": 0.77761114,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.79947972,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.769777297973633
    },
    {
      "auxiliary_loss_clip": 0.0115858,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 0.93136919,
      "balance_loss_mlp": 1.01767206,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 2.4261504083691308,
      "language_loss": 0.74822974,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.77007186,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 2.809373378753662
    },
    {
      "auxiliary_loss_clip": 0.01157433,
      "auxiliary_loss_mlp": 0.01024994,
      "balance_loss_clip": 0.96994001,
      "balance_loss_mlp": 1.01788652,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 1.9996893360988823,
      "language_loss": 0.82517308,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84699732,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 2.6016438007354736
    },
    {
      "auxiliary_loss_clip": 0.01172013,
      "auxiliary_loss_mlp": 0.01025112,
      "balance_loss_clip": 0.97121668,
      "balance_loss_mlp": 1.01776838,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 1.7679179105612366,
      "language_loss": 0.80866003,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.83063126,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 2.770411491394043
    },
    {
      "auxiliary_loss_clip": 0.01167062,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 0.93016315,
      "balance_loss_mlp": 1.020051,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 1.6446660946624412,
      "language_loss": 0.74573696,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.76767719,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.66302490234375
    },
    {
      "auxiliary_loss_clip": 0.01166867,
      "auxiliary_loss_mlp": 0.01122497,
      "balance_loss_clip": 1.01102984,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 1.8429998781418155,
      "language_loss": 0.76520216,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.78809577,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.6629695892333984
    },
    {
      "auxiliary_loss_clip": 0.01167213,
      "auxiliary_loss_mlp": 0.01024385,
      "balance_loss_clip": 1.01214838,
      "balance_loss_mlp": 1.01678574,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 1.9887281353594088,
      "language_loss": 0.68976521,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71168119,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 2.7103476524353027
    },
    {
      "auxiliary_loss_clip": 0.01170575,
      "auxiliary_loss_mlp": 0.01025118,
      "balance_loss_clip": 0.97305834,
      "balance_loss_mlp": 1.01768577,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 2.1180710734503307,
      "language_loss": 0.77719653,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.79915345,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.63236141204834
    },
    {
      "auxiliary_loss_clip": 0.01171703,
      "auxiliary_loss_mlp": 0.01023445,
      "balance_loss_clip": 1.01270914,
      "balance_loss_mlp": 1.01615286,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 3.9086473368845294,
      "language_loss": 0.76344866,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78540015,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 2.603902578353882
    },
    {
      "auxiliary_loss_clip": 0.01166925,
      "auxiliary_loss_mlp": 0.01021543,
      "balance_loss_clip": 1.04825175,
      "balance_loss_mlp": 1.0147686,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 1.9107756498221382,
      "language_loss": 0.85434079,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.87622541,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.573903799057007
    },
    {
      "auxiliary_loss_clip": 0.01162889,
      "auxiliary_loss_mlp": 0.01023957,
      "balance_loss_clip": 0.92924237,
      "balance_loss_mlp": 1.01684022,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.2947012553314177,
      "language_loss": 0.81654602,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.83841443,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 2.709505796432495
    },
    {
      "auxiliary_loss_clip": 0.01165686,
      "auxiliary_loss_mlp": 0.01028735,
      "balance_loss_clip": 0.97296512,
      "balance_loss_mlp": 1.02200532,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 1.6735393558039082,
      "language_loss": 0.67340082,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.69534504,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 2.7725701332092285
    },
    {
      "auxiliary_loss_clip": 0.01153215,
      "auxiliary_loss_mlp": 0.01033993,
      "balance_loss_clip": 0.8947888,
      "balance_loss_mlp": 1.02616715,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 1.669353679742519,
      "language_loss": 0.78054631,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80241841,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.7648751735687256
    },
    {
      "auxiliary_loss_clip": 0.01159364,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 0.92848909,
      "balance_loss_mlp": 1.01757574,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 1.6451645008632407,
      "language_loss": 0.78213352,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.80397487,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.6822903156280518
    },
    {
      "auxiliary_loss_clip": 0.01168748,
      "auxiliary_loss_mlp": 0.01022389,
      "balance_loss_clip": 0.93546486,
      "balance_loss_mlp": 1.0155139,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 2.6518496821640767,
      "language_loss": 0.71019197,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73210335,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 3.6434872150421143
    },
    {
      "auxiliary_loss_clip": 0.01159991,
      "auxiliary_loss_mlp": 0.01024717,
      "balance_loss_clip": 1.00939655,
      "balance_loss_mlp": 1.01791942,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 1.9391392617461072,
      "language_loss": 0.85683423,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.8786813,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 2.6505954265594482
    },
    {
      "auxiliary_loss_clip": 0.01171677,
      "auxiliary_loss_mlp": 0.01026119,
      "balance_loss_clip": 1.04949927,
      "balance_loss_mlp": 1.01835859,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 1.856408333513536,
      "language_loss": 0.72299093,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.74496889,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.57419490814209
    },
    {
      "auxiliary_loss_clip": 0.01162572,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 0.894068,
      "balance_loss_mlp": 1.01656926,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 1.9691145741919895,
      "language_loss": 0.75689489,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.778754,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.7622077465057373
    },
    {
      "auxiliary_loss_clip": 0.01169688,
      "auxiliary_loss_mlp": 0.01122273,
      "balance_loss_clip": 1.04949224,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 1.8624588089465899,
      "language_loss": 0.82385445,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.8467741,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 3.475412607192993
    },
    {
      "auxiliary_loss_clip": 0.01167423,
      "auxiliary_loss_mlp": 0.01024687,
      "balance_loss_clip": 1.01092553,
      "balance_loss_mlp": 1.01731944,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 2.26607066294627,
      "language_loss": 0.80081356,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.82273465,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 3.5931954383850098
    },
    {
      "auxiliary_loss_clip": 0.01168071,
      "auxiliary_loss_mlp": 0.01027126,
      "balance_loss_clip": 1.01056671,
      "balance_loss_mlp": 1.02040219,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 1.951561728586461,
      "language_loss": 0.83575368,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.85770571,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.631028413772583
    },
    {
      "auxiliary_loss_clip": 0.01168745,
      "auxiliary_loss_mlp": 0.01024732,
      "balance_loss_clip": 1.00935507,
      "balance_loss_mlp": 1.01782405,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.008962794131081,
      "language_loss": 0.84596169,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.8678965,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 3.5795249938964844
    },
    {
      "auxiliary_loss_clip": 0.01166162,
      "auxiliary_loss_mlp": 0.01026227,
      "balance_loss_clip": 1.01038003,
      "balance_loss_mlp": 1.0190115,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 1.9146542500653199,
      "language_loss": 0.78307998,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80500388,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.6702630519866943
    },
    {
      "auxiliary_loss_clip": 0.0116874,
      "auxiliary_loss_mlp": 0.01022097,
      "balance_loss_clip": 0.89670789,
      "balance_loss_mlp": 1.01524282,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.05980419626718,
      "language_loss": 0.74064565,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76255405,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.8144500255584717
    },
    {
      "auxiliary_loss_clip": 0.01162771,
      "auxiliary_loss_mlp": 0.01027063,
      "balance_loss_clip": 0.97054231,
      "balance_loss_mlp": 1.02008343,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 1.5480998905685954,
      "language_loss": 0.65695053,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.67884886,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.759544610977173
    },
    {
      "auxiliary_loss_clip": 0.01165504,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 0.93464828,
      "balance_loss_mlp": 1.01882935,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 2.6622261389536024,
      "language_loss": 0.80020392,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82211596,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.77209210395813
    },
    {
      "auxiliary_loss_clip": 0.01067276,
      "auxiliary_loss_mlp": 0.01004825,
      "balance_loss_clip": 0.9738338,
      "balance_loss_mlp": 1.00320339,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 0.8230380357023008,
      "language_loss": 0.55696952,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57769048,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 3.245089292526245
    },
    {
      "auxiliary_loss_clip": 0.0116806,
      "auxiliary_loss_mlp": 0.01025028,
      "balance_loss_clip": 0.97000754,
      "balance_loss_mlp": 1.01792336,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.7739860243954082,
      "language_loss": 0.75268686,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77461773,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 2.6758265495300293
    },
    {
      "auxiliary_loss_clip": 0.01165978,
      "auxiliary_loss_mlp": 0.01028006,
      "balance_loss_clip": 1.0478363,
      "balance_loss_mlp": 1.02111614,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 1.6832002587222736,
      "language_loss": 0.70866865,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.73060846,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 2.584362268447876
    },
    {
      "auxiliary_loss_clip": 0.01162645,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 0.97077096,
      "balance_loss_mlp": 1.02014506,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 1.7812504650415037,
      "language_loss": 0.69391561,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71581447,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.9021215438842773
    },
    {
      "auxiliary_loss_clip": 0.01176248,
      "auxiliary_loss_mlp": 0.01020334,
      "balance_loss_clip": 0.97412455,
      "balance_loss_mlp": 1.01222181,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 2.0518955951540114,
      "language_loss": 0.81634498,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.83831084,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.719414710998535
    },
    {
      "auxiliary_loss_clip": 0.01161489,
      "auxiliary_loss_mlp": 0.0102916,
      "balance_loss_clip": 0.97073066,
      "balance_loss_mlp": 1.02237999,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.6023095489972476,
      "language_loss": 0.81541294,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.83731949,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.6929712295532227
    },
    {
      "auxiliary_loss_clip": 0.01163338,
      "auxiliary_loss_mlp": 0.01024215,
      "balance_loss_clip": 1.00889397,
      "balance_loss_mlp": 1.01745248,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.414981776699237,
      "language_loss": 0.83597171,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.85784733,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 2.63899564743042
    },
    {
      "auxiliary_loss_clip": 0.01165731,
      "auxiliary_loss_mlp": 0.01025233,
      "balance_loss_clip": 0.89570057,
      "balance_loss_mlp": 1.01812828,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 2.1721276738499276,
      "language_loss": 0.73747623,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.75938594,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.72147536277771
    },
    {
      "auxiliary_loss_clip": 0.01157185,
      "auxiliary_loss_mlp": 0.01026484,
      "balance_loss_clip": 0.93214858,
      "balance_loss_mlp": 1.01858044,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.7850490025820416,
      "language_loss": 0.65425915,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.67609584,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.686843156814575
    },
    {
      "auxiliary_loss_clip": 0.01172424,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 0.97402745,
      "balance_loss_mlp": 1.01578236,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 1.540805114909941,
      "language_loss": 0.80414045,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82609808,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 2.7037429809570312
    },
    {
      "auxiliary_loss_clip": 0.01168026,
      "auxiliary_loss_mlp": 0.01029712,
      "balance_loss_clip": 1.01020074,
      "balance_loss_mlp": 1.02207077,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 2.2200653230471055,
      "language_loss": 0.83440137,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.85637873,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.681079149246216
    },
    {
      "auxiliary_loss_clip": 0.0108254,
      "auxiliary_loss_mlp": 0.01000486,
      "balance_loss_clip": 0.91460752,
      "balance_loss_mlp": 0.99872202,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.6837157235760004,
      "language_loss": 0.49621826,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51704854,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 3.428037643432617
    },
    {
      "auxiliary_loss_clip": 0.01164551,
      "auxiliary_loss_mlp": 0.0102228,
      "balance_loss_clip": 0.85834384,
      "balance_loss_mlp": 1.01468015,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 1.869122991605269,
      "language_loss": 0.71968865,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.741557,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 2.8172593116760254
    },
    {
      "auxiliary_loss_clip": 0.01171795,
      "auxiliary_loss_mlp": 0.01027332,
      "balance_loss_clip": 1.05072665,
      "balance_loss_mlp": 1.02045059,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 2.1434488149965705,
      "language_loss": 0.88239866,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90438992,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.995695114135742
    },
    {
      "auxiliary_loss_clip": 0.01165406,
      "auxiliary_loss_mlp": 0.01024633,
      "balance_loss_clip": 0.93188691,
      "balance_loss_mlp": 1.01804626,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.6143880129196404,
      "language_loss": 0.74409837,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76599872,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 3.5115628242492676
    },
    {
      "auxiliary_loss_clip": 0.01163749,
      "auxiliary_loss_mlp": 0.01025482,
      "balance_loss_clip": 0.97494304,
      "balance_loss_mlp": 1.01868963,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 1.8142404141276343,
      "language_loss": 0.71878338,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.74067575,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 2.7625370025634766
    },
    {
      "auxiliary_loss_clip": 0.01164959,
      "auxiliary_loss_mlp": 0.0102609,
      "balance_loss_clip": 0.93418717,
      "balance_loss_mlp": 1.01877391,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 1.7206305437124898,
      "language_loss": 0.68538928,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.70729971,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 2.732185125350952
    },
    {
      "auxiliary_loss_clip": 0.01063427,
      "auxiliary_loss_mlp": 0.01002404,
      "balance_loss_clip": 0.93563545,
      "balance_loss_mlp": 1.00071132,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 0.9596675388848459,
      "language_loss": 0.56813359,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58879191,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.322251796722412
    },
    {
      "auxiliary_loss_clip": 0.01168475,
      "auxiliary_loss_mlp": 0.01028619,
      "balance_loss_clip": 0.975986,
      "balance_loss_mlp": 1.0204885,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 2.230171267996136,
      "language_loss": 0.77289867,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79486954,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 3.7444915771484375
    },
    {
      "auxiliary_loss_clip": 0.01160627,
      "auxiliary_loss_mlp": 0.01024232,
      "balance_loss_clip": 1.00795388,
      "balance_loss_mlp": 1.0171299,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 1.7217557232560954,
      "language_loss": 0.72844744,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.75029606,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 3.6195614337921143
    },
    {
      "auxiliary_loss_clip": 0.01166564,
      "auxiliary_loss_mlp": 0.01029088,
      "balance_loss_clip": 0.93152708,
      "balance_loss_mlp": 1.02197695,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.4822379584571816,
      "language_loss": 0.78396606,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80592251,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.7006921768188477
    },
    {
      "auxiliary_loss_clip": 0.01169895,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.01015425,
      "balance_loss_mlp": 1.01905882,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 1.8379066023896298,
      "language_loss": 0.71785706,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.73982346,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 3.575958013534546
    },
    {
      "auxiliary_loss_clip": 0.01149313,
      "auxiliary_loss_mlp": 0.01122555,
      "balance_loss_clip": 0.85397136,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 1.6383942505293292,
      "language_loss": 0.75755298,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.78027165,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.815039873123169
    },
    {
      "auxiliary_loss_clip": 0.01162385,
      "auxiliary_loss_mlp": 0.01023549,
      "balance_loss_clip": 1.04665256,
      "balance_loss_mlp": 1.0170399,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 1.8377371701805054,
      "language_loss": 0.65660334,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67846268,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.6383023262023926
    },
    {
      "auxiliary_loss_clip": 0.011677,
      "auxiliary_loss_mlp": 0.0103079,
      "balance_loss_clip": 1.04746413,
      "balance_loss_mlp": 1.02282727,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.6286177910948578,
      "language_loss": 0.85206056,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87404549,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.593885660171509
    },
    {
      "auxiliary_loss_clip": 0.0117231,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 0.93691242,
      "balance_loss_mlp": 1.01779509,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 1.6336485903650038,
      "language_loss": 0.83221173,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85417998,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.718355655670166
    },
    {
      "auxiliary_loss_clip": 0.0116235,
      "auxiliary_loss_mlp": 0.01021622,
      "balance_loss_clip": 1.00880098,
      "balance_loss_mlp": 1.01519632,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 1.6270422691171067,
      "language_loss": 0.66173309,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68357283,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 2.6259047985076904
    },
    {
      "auxiliary_loss_clip": 0.01063324,
      "auxiliary_loss_mlp": 0.01003477,
      "balance_loss_clip": 1.01040363,
      "balance_loss_mlp": 1.00190377,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.21937876737063,
      "language_loss": 0.67847002,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69913805,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 3.011660099029541
    },
    {
      "auxiliary_loss_clip": 0.01158836,
      "auxiliary_loss_mlp": 0.01022333,
      "balance_loss_clip": 0.97234547,
      "balance_loss_mlp": 1.01551425,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 1.8851762898704747,
      "language_loss": 0.75876689,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78057861,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 2.690715789794922
    },
    {
      "auxiliary_loss_clip": 0.01067776,
      "auxiliary_loss_mlp": 0.01002058,
      "balance_loss_clip": 0.97489661,
      "balance_loss_mlp": 1.00047302,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7422208074566918,
      "language_loss": 0.56463599,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58533442,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 3.0959746837615967
    },
    {
      "auxiliary_loss_clip": 0.01163368,
      "auxiliary_loss_mlp": 0.01024483,
      "balance_loss_clip": 1.0067327,
      "balance_loss_mlp": 1.0177598,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 1.8361782769474089,
      "language_loss": 0.85136926,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.87324774,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 2.6695306301116943
    },
    {
      "auxiliary_loss_clip": 0.01165834,
      "auxiliary_loss_mlp": 0.0102266,
      "balance_loss_clip": 0.93178445,
      "balance_loss_mlp": 1.0155549,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 1.9157808523326956,
      "language_loss": 0.69396603,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.71585101,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.7274863719940186
    },
    {
      "auxiliary_loss_clip": 0.01167719,
      "auxiliary_loss_mlp": 0.01024118,
      "balance_loss_clip": 1.04803598,
      "balance_loss_mlp": 1.01740336,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 2.0568954607560443,
      "language_loss": 0.77170408,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79362249,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 2.6576554775238037
    },
    {
      "auxiliary_loss_clip": 0.01163053,
      "auxiliary_loss_mlp": 0.01027529,
      "balance_loss_clip": 0.97049344,
      "balance_loss_mlp": 1.02082682,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 2.4124370433303737,
      "language_loss": 0.81402707,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83593285,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.692084550857544
    },
    {
      "auxiliary_loss_clip": 0.01168924,
      "auxiliary_loss_mlp": 0.01023949,
      "balance_loss_clip": 0.97274852,
      "balance_loss_mlp": 1.01669538,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.1642840333851705,
      "language_loss": 0.70782042,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.7297492,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.7064127922058105
    },
    {
      "auxiliary_loss_clip": 0.01162322,
      "auxiliary_loss_mlp": 0.01020744,
      "balance_loss_clip": 1.00814641,
      "balance_loss_mlp": 1.01343989,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 1.9396684732344072,
      "language_loss": 0.7402516,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.76208228,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 2.682731866836548
    },
    {
      "auxiliary_loss_clip": 0.01169014,
      "auxiliary_loss_mlp": 0.0103073,
      "balance_loss_clip": 0.97068763,
      "balance_loss_mlp": 1.02354491,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 1.853263914336021,
      "language_loss": 0.73238635,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.7543838,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.744798421859741
    },
    {
      "auxiliary_loss_clip": 0.01166357,
      "auxiliary_loss_mlp": 0.01027081,
      "balance_loss_clip": 0.97145832,
      "balance_loss_mlp": 1.0204227,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.4653408911671355,
      "language_loss": 0.79966152,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82159591,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 2.642831802368164
    },
    {
      "auxiliary_loss_clip": 0.0115842,
      "auxiliary_loss_mlp": 0.01030314,
      "balance_loss_clip": 1.00866055,
      "balance_loss_mlp": 1.02315891,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 2.1180891724067004,
      "language_loss": 0.84771603,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.8696034,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.694859504699707
    },
    {
      "auxiliary_loss_clip": 0.01161551,
      "auxiliary_loss_mlp": 0.01028605,
      "balance_loss_clip": 1.00866771,
      "balance_loss_mlp": 1.02067471,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.5340970447865012,
      "language_loss": 0.72729015,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.74919164,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 2.6800050735473633
    },
    {
      "auxiliary_loss_clip": 0.01168001,
      "auxiliary_loss_mlp": 0.01032157,
      "balance_loss_clip": 1.05068648,
      "balance_loss_mlp": 1.02515614,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 1.6936884632718527,
      "language_loss": 0.74596763,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.76796925,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 3.5063555240631104
    },
    {
      "auxiliary_loss_clip": 0.01164117,
      "auxiliary_loss_mlp": 0.01026638,
      "balance_loss_clip": 0.97361779,
      "balance_loss_mlp": 1.01970601,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 1.9232689236137521,
      "language_loss": 0.79533905,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81724656,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.6762590408325195
    },
    {
      "auxiliary_loss_clip": 0.01166742,
      "auxiliary_loss_mlp": 0.01026468,
      "balance_loss_clip": 1.00993729,
      "balance_loss_mlp": 1.0195694,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 2.010851984329852,
      "language_loss": 0.70516449,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.72709656,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 2.610081195831299
    },
    {
      "auxiliary_loss_clip": 0.01161885,
      "auxiliary_loss_mlp": 0.01021065,
      "balance_loss_clip": 0.9703151,
      "balance_loss_mlp": 1.01435399,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 1.4802200120002522,
      "language_loss": 0.82225919,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84408867,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 2.698878288269043
    },
    {
      "auxiliary_loss_clip": 0.01172361,
      "auxiliary_loss_mlp": 0.01019026,
      "balance_loss_clip": 0.97030413,
      "balance_loss_mlp": 1.01152444,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 3.1983074497222344,
      "language_loss": 0.72324759,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.74516147,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 3.5795304775238037
    },
    {
      "auxiliary_loss_clip": 0.01165587,
      "auxiliary_loss_mlp": 0.01025601,
      "balance_loss_clip": 0.96907341,
      "balance_loss_mlp": 1.01924443,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.3086744408043325,
      "language_loss": 0.73002321,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.75193506,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 3.4847850799560547
    },
    {
      "auxiliary_loss_clip": 0.01163769,
      "auxiliary_loss_mlp": 0.01021764,
      "balance_loss_clip": 0.93005741,
      "balance_loss_mlp": 1.01531446,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.5299940507318315,
      "language_loss": 0.75844264,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78029799,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.728302240371704
    },
    {
      "auxiliary_loss_clip": 0.01173595,
      "auxiliary_loss_mlp": 0.01026595,
      "balance_loss_clip": 0.97135437,
      "balance_loss_mlp": 1.01952004,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 1.6278239716917,
      "language_loss": 0.76379788,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.7857998,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 3.5740838050842285
    },
    {
      "auxiliary_loss_clip": 0.0116427,
      "auxiliary_loss_mlp": 0.01027522,
      "balance_loss_clip": 0.96895826,
      "balance_loss_mlp": 1.02039671,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 9.138084926426746,
      "language_loss": 0.75875431,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.78067225,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.6749520301818848
    },
    {
      "auxiliary_loss_clip": 0.01167717,
      "auxiliary_loss_mlp": 0.01025936,
      "balance_loss_clip": 1.00795841,
      "balance_loss_mlp": 1.01851833,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 2.095903663586103,
      "language_loss": 0.75387985,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.77581638,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.672379493713379
    },
    {
      "auxiliary_loss_clip": 0.01163152,
      "auxiliary_loss_mlp": 0.01027034,
      "balance_loss_clip": 0.89209306,
      "balance_loss_mlp": 1.0197922,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.345147455211459,
      "language_loss": 0.85794044,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.87984228,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.7379651069641113
    },
    {
      "auxiliary_loss_clip": 0.01168667,
      "auxiliary_loss_mlp": 0.01028737,
      "balance_loss_clip": 1.00909352,
      "balance_loss_mlp": 1.02122068,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.9092164255475486,
      "language_loss": 0.68976581,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.7117399,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.693866014480591
    },
    {
      "auxiliary_loss_clip": 0.01160351,
      "auxiliary_loss_mlp": 0.010239,
      "balance_loss_clip": 1.0069294,
      "balance_loss_mlp": 1.0173974,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.579703520937219,
      "language_loss": 0.69695044,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.71879292,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 2.65807843208313
    },
    {
      "auxiliary_loss_clip": 0.01169583,
      "auxiliary_loss_mlp": 0.01123116,
      "balance_loss_clip": 0.9724232,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 3.2331822222322684,
      "language_loss": 0.82194078,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.84486777,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.6998610496520996
    },
    {
      "auxiliary_loss_clip": 0.01175457,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 0.93493885,
      "balance_loss_mlp": 1.02085102,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 2.5333383256226725,
      "language_loss": 0.77645075,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79849178,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 2.7152202129364014
    },
    {
      "auxiliary_loss_clip": 0.01167928,
      "auxiliary_loss_mlp": 0.01030931,
      "balance_loss_clip": 0.96816015,
      "balance_loss_mlp": 1.02370667,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 1.8868751664829704,
      "language_loss": 0.81698221,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.83897078,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.7446658611297607
    },
    {
      "auxiliary_loss_clip": 0.01161924,
      "auxiliary_loss_mlp": 0.01122083,
      "balance_loss_clip": 0.85581923,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 1.9364528251360784,
      "language_loss": 0.64587569,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.66871578,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 2.828670024871826
    },
    {
      "auxiliary_loss_clip": 0.0117289,
      "auxiliary_loss_mlp": 0.01026581,
      "balance_loss_clip": 0.93401462,
      "balance_loss_mlp": 1.01927686,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 1.547993456622944,
      "language_loss": 0.8146143,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83660901,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 2.7291746139526367
    },
    {
      "auxiliary_loss_clip": 0.01170117,
      "auxiliary_loss_mlp": 0.01024571,
      "balance_loss_clip": 0.97216326,
      "balance_loss_mlp": 1.01695323,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 2.2248979435499234,
      "language_loss": 0.80648279,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.82842964,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 2.748835802078247
    },
    {
      "auxiliary_loss_clip": 0.0117129,
      "auxiliary_loss_mlp": 0.01022401,
      "balance_loss_clip": 0.97113931,
      "balance_loss_mlp": 1.01491427,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 6.576434311991205,
      "language_loss": 0.7516712,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77360809,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.7035574913024902
    },
    {
      "auxiliary_loss_clip": 0.01161866,
      "auxiliary_loss_mlp": 0.0102204,
      "balance_loss_clip": 0.93378186,
      "balance_loss_mlp": 1.01503658,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.306189732970349,
      "language_loss": 0.72167337,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74351245,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.6859192848205566
    },
    {
      "auxiliary_loss_clip": 0.01157555,
      "auxiliary_loss_mlp": 0.01023401,
      "balance_loss_clip": 1.00565004,
      "balance_loss_mlp": 1.01659143,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.5737589753669832,
      "language_loss": 0.71080434,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.73261392,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.7008285522460938
    },
    {
      "auxiliary_loss_clip": 0.01161109,
      "auxiliary_loss_mlp": 0.01023036,
      "balance_loss_clip": 0.96811724,
      "balance_loss_mlp": 1.01579404,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 1.7872178149622389,
      "language_loss": 0.78243166,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.80427307,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.6156487464904785
    },
    {
      "auxiliary_loss_clip": 0.011713,
      "auxiliary_loss_mlp": 0.01023939,
      "balance_loss_clip": 0.93432033,
      "balance_loss_mlp": 1.01705432,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 2.442105850186843,
      "language_loss": 0.75531763,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.77726996,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.7446305751800537
    },
    {
      "auxiliary_loss_clip": 0.01167331,
      "auxiliary_loss_mlp": 0.01029933,
      "balance_loss_clip": 1.0119884,
      "balance_loss_mlp": 1.02239048,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 1.7694075115171133,
      "language_loss": 0.78568983,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.80766249,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 2.688480854034424
    },
    {
      "auxiliary_loss_clip": 0.01160754,
      "auxiliary_loss_mlp": 0.01027364,
      "balance_loss_clip": 1.00915575,
      "balance_loss_mlp": 1.02037215,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 1.929614902855729,
      "language_loss": 0.79737502,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.81925625,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 2.628587007522583
    },
    {
      "auxiliary_loss_clip": 0.01163387,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.01020133,
      "balance_loss_mlp": 1.01797414,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 1.583104105760063,
      "language_loss": 0.7695998,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.79148126,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 3.557494878768921
    },
    {
      "auxiliary_loss_clip": 0.01161582,
      "auxiliary_loss_mlp": 0.01024224,
      "balance_loss_clip": 0.97200125,
      "balance_loss_mlp": 1.01744699,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.390093888177149,
      "language_loss": 0.73339248,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.75525045,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 2.6819956302642822
    },
    {
      "auxiliary_loss_clip": 0.0116332,
      "auxiliary_loss_mlp": 0.01022403,
      "balance_loss_clip": 0.971434,
      "balance_loss_mlp": 1.01599872,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 1.393954668526962,
      "language_loss": 0.79511148,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.8169688,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 2.686934471130371
    },
    {
      "auxiliary_loss_clip": 0.01167504,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 0.97096717,
      "balance_loss_mlp": 1.01903105,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 1.9730534198418257,
      "language_loss": 0.8324827,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85441601,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.6666321754455566
    },
    {
      "auxiliary_loss_clip": 0.01072282,
      "auxiliary_loss_mlp": 0.01000518,
      "balance_loss_clip": 0.93748438,
      "balance_loss_mlp": 0.99870569,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7411716958981042,
      "language_loss": 0.56125093,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58197892,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 4.170970916748047
    },
    {
      "auxiliary_loss_clip": 0.01161055,
      "auxiliary_loss_mlp": 0.0102264,
      "balance_loss_clip": 0.85514885,
      "balance_loss_mlp": 1.01570511,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 3.0958321074483557,
      "language_loss": 0.76188886,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78372586,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 2.8347702026367188
    },
    {
      "auxiliary_loss_clip": 0.01162441,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 0.97118378,
      "balance_loss_mlp": 1.01846826,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 1.7800399936595284,
      "language_loss": 0.61192757,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63380694,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 3.630631685256958
    },
    {
      "auxiliary_loss_clip": 0.01164667,
      "auxiliary_loss_mlp": 0.01028459,
      "balance_loss_clip": 0.93299437,
      "balance_loss_mlp": 1.0205884,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 1.6964526386881988,
      "language_loss": 0.68297446,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.70490569,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.718451976776123
    },
    {
      "auxiliary_loss_clip": 0.01171601,
      "auxiliary_loss_mlp": 0.01028565,
      "balance_loss_clip": 1.00860274,
      "balance_loss_mlp": 1.02146578,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 1.7353813465537344,
      "language_loss": 0.6574645,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.67946613,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.6427536010742188
    },
    {
      "auxiliary_loss_clip": 0.01170381,
      "auxiliary_loss_mlp": 0.01021023,
      "balance_loss_clip": 1.00899541,
      "balance_loss_mlp": 1.01412106,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 2.073061422444826,
      "language_loss": 0.69489336,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71680743,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.6187357902526855
    },
    {
      "auxiliary_loss_clip": 0.01170571,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.00848472,
      "balance_loss_mlp": 1.02126288,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 2.0622248769270044,
      "language_loss": 0.81465399,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.83664453,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.6951143741607666
    },
    {
      "auxiliary_loss_clip": 0.01061691,
      "auxiliary_loss_mlp": 0.01005176,
      "balance_loss_clip": 0.93614584,
      "balance_loss_mlp": 1.00367379,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.820956838326652,
      "language_loss": 0.62007618,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64074486,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.326580286026001
    },
    {
      "auxiliary_loss_clip": 0.01167118,
      "auxiliary_loss_mlp": 0.01025134,
      "balance_loss_clip": 1.01094913,
      "balance_loss_mlp": 1.01807714,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 2.0315784629572753,
      "language_loss": 0.69702351,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.71894604,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 2.6353323459625244
    },
    {
      "auxiliary_loss_clip": 0.01163067,
      "auxiliary_loss_mlp": 0.01023718,
      "balance_loss_clip": 0.97034329,
      "balance_loss_mlp": 1.01720941,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.8352351794439536,
      "language_loss": 0.72477031,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74663818,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.6558969020843506
    },
    {
      "auxiliary_loss_clip": 0.01068188,
      "auxiliary_loss_mlp": 0.01001167,
      "balance_loss_clip": 0.97388166,
      "balance_loss_mlp": 0.99955809,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 0.7795546180626605,
      "language_loss": 0.61085325,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.6315468,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 3.156362533569336
    },
    {
      "auxiliary_loss_clip": 0.01153814,
      "auxiliary_loss_mlp": 0.01022993,
      "balance_loss_clip": 0.93021196,
      "balance_loss_mlp": 1.01589739,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 1.8170139052768077,
      "language_loss": 0.84814858,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.86991668,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.7334983348846436
    },
    {
      "auxiliary_loss_clip": 0.01169743,
      "auxiliary_loss_mlp": 0.01025201,
      "balance_loss_clip": 1.04903948,
      "balance_loss_mlp": 1.01765776,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 1.7387485494983403,
      "language_loss": 0.78921723,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.8111667,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.6530768871307373
    },
    {
      "auxiliary_loss_clip": 0.01165416,
      "auxiliary_loss_mlp": 0.01025831,
      "balance_loss_clip": 1.04711974,
      "balance_loss_mlp": 1.01905417,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 2.465520945436477,
      "language_loss": 0.79990935,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.82182181,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 2.5989184379577637
    },
    {
      "auxiliary_loss_clip": 0.01156929,
      "auxiliary_loss_mlp": 0.01122797,
      "balance_loss_clip": 0.89354134,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.52534927573421,
      "language_loss": 0.71548325,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73828053,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 2.7455480098724365
    },
    {
      "auxiliary_loss_clip": 0.01155453,
      "auxiliary_loss_mlp": 0.01025509,
      "balance_loss_clip": 0.93091583,
      "balance_loss_mlp": 1.01922929,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 1.8428465720108593,
      "language_loss": 0.71337295,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73518258,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.697756290435791
    },
    {
      "auxiliary_loss_clip": 0.01171151,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.05121911,
      "balance_loss_mlp": 1.01980352,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 1.9679518759197403,
      "language_loss": 0.82279021,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.84476608,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.6269376277923584
    },
    {
      "auxiliary_loss_clip": 0.01159608,
      "auxiliary_loss_mlp": 0.01122889,
      "balance_loss_clip": 0.92986596,
      "balance_loss_mlp": 0.0,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.18788593387655,
      "language_loss": 0.76658511,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.78941011,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.7823903560638428
    },
    {
      "auxiliary_loss_clip": 0.01179434,
      "auxiliary_loss_mlp": 0.01029022,
      "balance_loss_clip": 0.89691222,
      "balance_loss_mlp": 1.02175021,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 2.031269667282568,
      "language_loss": 0.81127858,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83336306,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.7369227409362793
    },
    {
      "auxiliary_loss_clip": 0.01175532,
      "auxiliary_loss_mlp": 0.01029268,
      "balance_loss_clip": 0.97309375,
      "balance_loss_mlp": 1.02176952,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 2.6429454255930698,
      "language_loss": 0.76447153,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.78651953,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.6901795864105225
    },
    {
      "auxiliary_loss_clip": 0.01169845,
      "auxiliary_loss_mlp": 0.01025015,
      "balance_loss_clip": 0.97130704,
      "balance_loss_mlp": 1.01771367,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 1.9984939830232356,
      "language_loss": 0.72444832,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.7463969,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 2.7669506072998047
    },
    {
      "auxiliary_loss_clip": 0.01163109,
      "auxiliary_loss_mlp": 0.0102877,
      "balance_loss_clip": 0.97072452,
      "balance_loss_mlp": 1.02188563,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 1.8062776195677566,
      "language_loss": 0.74932498,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.77124369,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 2.723005533218384
    },
    {
      "auxiliary_loss_clip": 0.01157873,
      "auxiliary_loss_mlp": 0.01023037,
      "balance_loss_clip": 0.93037558,
      "balance_loss_mlp": 1.01563394,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 2.0988520743212282,
      "language_loss": 0.83620346,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.85801256,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 3.668250322341919
    },
    {
      "auxiliary_loss_clip": 0.01168025,
      "auxiliary_loss_mlp": 0.01021842,
      "balance_loss_clip": 1.04880643,
      "balance_loss_mlp": 1.01502001,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 1.9398632161052807,
      "language_loss": 0.76821017,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.7901088,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.5816712379455566
    },
    {
      "auxiliary_loss_clip": 0.01154335,
      "auxiliary_loss_mlp": 0.01025157,
      "balance_loss_clip": 0.85330981,
      "balance_loss_mlp": 1.0177573,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 1.8208081024151939,
      "language_loss": 0.71439797,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.73619294,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.8015496730804443
    },
    {
      "auxiliary_loss_clip": 0.01167513,
      "auxiliary_loss_mlp": 0.01029335,
      "balance_loss_clip": 0.81934023,
      "balance_loss_mlp": 1.02175951,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 1.998100647694979,
      "language_loss": 0.64877552,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.670744,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 2.7681431770324707
    },
    {
      "auxiliary_loss_clip": 0.01174565,
      "auxiliary_loss_mlp": 0.01028505,
      "balance_loss_clip": 0.97725439,
      "balance_loss_mlp": 1.02042818,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.5260398790855778,
      "language_loss": 0.64301598,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66504669,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 3.865255117416382
    },
    {
      "auxiliary_loss_clip": 0.01167636,
      "auxiliary_loss_mlp": 0.01025484,
      "balance_loss_clip": 1.01027632,
      "balance_loss_mlp": 1.01820004,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 2.7486408435464105,
      "language_loss": 0.69040704,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71233821,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.6618881225585938
    },
    {
      "auxiliary_loss_clip": 0.01152618,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 0.93272954,
      "balance_loss_mlp": 1.02027822,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 1.6985888299093053,
      "language_loss": 0.80291706,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82472062,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 3.6050167083740234
    },
    {
      "auxiliary_loss_clip": 0.0116247,
      "auxiliary_loss_mlp": 0.01022999,
      "balance_loss_clip": 1.00658488,
      "balance_loss_mlp": 1.01592422,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 3.9578539963081094,
      "language_loss": 0.74832761,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.77018225,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.6884853839874268
    },
    {
      "auxiliary_loss_clip": 0.01159187,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 0.97060877,
      "balance_loss_mlp": 1.02157116,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.1263267032403146,
      "language_loss": 0.76973987,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.79161882,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.668391704559326
    },
    {
      "auxiliary_loss_clip": 0.01167288,
      "auxiliary_loss_mlp": 0.01023631,
      "balance_loss_clip": 1.04858565,
      "balance_loss_mlp": 1.01683652,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 2.122286649202501,
      "language_loss": 0.76322603,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78513527,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.697758197784424
    },
    {
      "auxiliary_loss_clip": 0.01169164,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 0.97593999,
      "balance_loss_mlp": 1.01845157,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 1.8647662892298509,
      "language_loss": 0.77482504,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.79677588,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.686018466949463
    },
    {
      "auxiliary_loss_clip": 0.01158882,
      "auxiliary_loss_mlp": 0.01025095,
      "balance_loss_clip": 1.00829852,
      "balance_loss_mlp": 1.01817179,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.536989146752822,
      "language_loss": 0.73175621,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.75359595,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.70660662651062
    },
    {
      "auxiliary_loss_clip": 0.01082657,
      "auxiliary_loss_mlp": 0.01000842,
      "balance_loss_clip": 0.91338211,
      "balance_loss_mlp": 0.99902958,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8340547015158464,
      "language_loss": 0.55015856,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.57099354,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 3.2129435539245605
    },
    {
      "auxiliary_loss_clip": 0.01162205,
      "auxiliary_loss_mlp": 0.01030174,
      "balance_loss_clip": 0.8965385,
      "balance_loss_mlp": 1.02288175,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 2.2155370838604886,
      "language_loss": 0.80981565,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.83173943,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.695000410079956
    },
    {
      "auxiliary_loss_clip": 0.01161196,
      "auxiliary_loss_mlp": 0.01122178,
      "balance_loss_clip": 0.93440074,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 2.0095648123742302,
      "language_loss": 0.71525902,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.73809278,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 2.774130344390869
    },
    {
      "auxiliary_loss_clip": 0.01172107,
      "auxiliary_loss_mlp": 0.01027607,
      "balance_loss_clip": 1.01128829,
      "balance_loss_mlp": 1.02022469,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 1.7777227356774268,
      "language_loss": 0.81793553,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.83993268,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.6635100841522217
    },
    {
      "auxiliary_loss_clip": 0.01176584,
      "auxiliary_loss_mlp": 0.01035369,
      "balance_loss_clip": 1.05149722,
      "balance_loss_mlp": 1.02744722,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.2488770822622866,
      "language_loss": 0.70033896,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72245848,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.7339258193969727
    },
    {
      "auxiliary_loss_clip": 0.01154532,
      "auxiliary_loss_mlp": 0.01030434,
      "balance_loss_clip": 0.93122947,
      "balance_loss_mlp": 1.02314115,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 1.6966805971671244,
      "language_loss": 0.81552249,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83737212,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 2.691976547241211
    },
    {
      "auxiliary_loss_clip": 0.0117242,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 0.97352171,
      "balance_loss_mlp": 1.02251744,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.6600458766294686,
      "language_loss": 0.6250807,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64709669,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 2.793504476547241
    },
    {
      "auxiliary_loss_clip": 0.01167164,
      "auxiliary_loss_mlp": 0.01021576,
      "balance_loss_clip": 1.04716372,
      "balance_loss_mlp": 1.0146296,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 2.04525475400459,
      "language_loss": 0.75137782,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77326524,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.615560293197632
    },
    {
      "auxiliary_loss_clip": 0.01075192,
      "auxiliary_loss_mlp": 0.01004353,
      "balance_loss_clip": 0.89919889,
      "balance_loss_mlp": 1.00267208,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7925608282809694,
      "language_loss": 0.58605999,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.60685539,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.34809947013855
    },
    {
      "auxiliary_loss_clip": 0.01156199,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 0.96915317,
      "balance_loss_mlp": 1.01898193,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.417366871430993,
      "language_loss": 0.68307471,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70489395,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.7659945487976074
    },
    {
      "auxiliary_loss_clip": 0.01165146,
      "auxiliary_loss_mlp": 0.01023664,
      "balance_loss_clip": 0.97140998,
      "balance_loss_mlp": 1.01683605,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 2.3960918920804724,
      "language_loss": 0.8226186,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84450674,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.757293701171875
    },
    {
      "auxiliary_loss_clip": 0.01172018,
      "auxiliary_loss_mlp": 0.01022925,
      "balance_loss_clip": 0.93050599,
      "balance_loss_mlp": 1.01557565,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 2.429236493032102,
      "language_loss": 0.79713655,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.81908602,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.694286584854126
    },
    {
      "auxiliary_loss_clip": 0.01171462,
      "auxiliary_loss_mlp": 0.01023514,
      "balance_loss_clip": 1.05099893,
      "balance_loss_mlp": 1.01673102,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 1.9229153987363328,
      "language_loss": 0.81156933,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.8335191,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.641557455062866
    },
    {
      "auxiliary_loss_clip": 0.0116439,
      "auxiliary_loss_mlp": 0.01026608,
      "balance_loss_clip": 1.00910878,
      "balance_loss_mlp": 1.01981592,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 1.9453484554501685,
      "language_loss": 0.89136481,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.91327477,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 2.667766571044922
    },
    {
      "auxiliary_loss_clip": 0.01171564,
      "auxiliary_loss_mlp": 0.01024458,
      "balance_loss_clip": 1.04894078,
      "balance_loss_mlp": 1.01667619,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 2.620676653309956,
      "language_loss": 0.70235562,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.72431588,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 3.4663212299346924
    },
    {
      "auxiliary_loss_clip": 0.01168173,
      "auxiliary_loss_mlp": 0.01025147,
      "balance_loss_clip": 1.04680252,
      "balance_loss_mlp": 1.01731443,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 3.1698578003948525,
      "language_loss": 0.69904572,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.72097892,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 2.605949640274048
    },
    {
      "auxiliary_loss_clip": 0.01157786,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 0.96996009,
      "balance_loss_mlp": 1.01842844,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 1.971532283967838,
      "language_loss": 0.68016803,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70199895,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.676250696182251
    },
    {
      "auxiliary_loss_clip": 0.01167328,
      "auxiliary_loss_mlp": 0.01021382,
      "balance_loss_clip": 1.00962079,
      "balance_loss_mlp": 1.01477432,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 2.2040427960682725,
      "language_loss": 0.79869539,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.82058251,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.773226499557495
    },
    {
      "auxiliary_loss_clip": 0.01165557,
      "auxiliary_loss_mlp": 0.0102476,
      "balance_loss_clip": 1.01239276,
      "balance_loss_mlp": 1.01850486,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 1.753680853444614,
      "language_loss": 0.83837008,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.8602733,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 3.56089448928833
    },
    {
      "auxiliary_loss_clip": 0.01164085,
      "auxiliary_loss_mlp": 0.01122101,
      "balance_loss_clip": 0.97261274,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 1.826576486647448,
      "language_loss": 0.80132937,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.82419121,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.759669542312622
    },
    {
      "auxiliary_loss_clip": 0.01168724,
      "auxiliary_loss_mlp": 0.01026194,
      "balance_loss_clip": 1.04881227,
      "balance_loss_mlp": 1.01920223,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 1.692583213526655,
      "language_loss": 0.81071889,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83266807,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 3.5582456588745117
    },
    {
      "auxiliary_loss_clip": 0.01157149,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.00646126,
      "balance_loss_mlp": 1.01984155,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.51368885116395,
      "language_loss": 0.7391969,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.761033,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.8561441898345947
    },
    {
      "auxiliary_loss_clip": 0.01165504,
      "auxiliary_loss_mlp": 0.01029074,
      "balance_loss_clip": 1.04757333,
      "balance_loss_mlp": 1.02211785,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 2.1263647850678713,
      "language_loss": 0.66668785,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.68863362,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.62213134765625
    },
    {
      "auxiliary_loss_clip": 0.01158483,
      "auxiliary_loss_mlp": 0.01031978,
      "balance_loss_clip": 1.00456989,
      "balance_loss_mlp": 1.02478957,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.2498722394846977,
      "language_loss": 0.73120737,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.75311196,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.6561989784240723
    },
    {
      "auxiliary_loss_clip": 0.01171956,
      "auxiliary_loss_mlp": 0.01021595,
      "balance_loss_clip": 1.00901341,
      "balance_loss_mlp": 1.01494896,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 1.88807715175308,
      "language_loss": 0.76895911,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.79089463,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.6492528915405273
    },
    {
      "auxiliary_loss_clip": 0.01175311,
      "auxiliary_loss_mlp": 0.01027629,
      "balance_loss_clip": 1.01182973,
      "balance_loss_mlp": 1.01986897,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 2.042689631333684,
      "language_loss": 0.68758869,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.70961815,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 2.790853500366211
    },
    {
      "auxiliary_loss_clip": 0.01168519,
      "auxiliary_loss_mlp": 0.01022512,
      "balance_loss_clip": 1.0488987,
      "balance_loss_mlp": 1.01547027,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.9173437882851636,
      "language_loss": 0.83792275,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.85983312,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 2.5876197814941406
    },
    {
      "auxiliary_loss_clip": 0.01150113,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 0.96605456,
      "balance_loss_mlp": 1.021878,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 1.6263011755623462,
      "language_loss": 0.75985128,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.7816478,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.8076512813568115
    },
    {
      "auxiliary_loss_clip": 0.01164105,
      "auxiliary_loss_mlp": 0.01026098,
      "balance_loss_clip": 0.97162163,
      "balance_loss_mlp": 1.01962507,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 2.088237360631757,
      "language_loss": 0.78823125,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.81013328,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 2.7240779399871826
    },
    {
      "auxiliary_loss_clip": 0.01168545,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 0.93392229,
      "balance_loss_mlp": 1.01837087,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 2.1330679735076217,
      "language_loss": 0.80009407,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82203197,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.781064748764038
    },
    {
      "auxiliary_loss_clip": 0.01161599,
      "auxiliary_loss_mlp": 0.01028136,
      "balance_loss_clip": 0.89363968,
      "balance_loss_mlp": 1.02042866,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 1.9874740497266743,
      "language_loss": 0.81518251,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.83707988,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.730647325515747
    },
    {
      "auxiliary_loss_clip": 0.01161758,
      "auxiliary_loss_mlp": 0.01028839,
      "balance_loss_clip": 1.00844634,
      "balance_loss_mlp": 1.0217247,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 2.1071942755616333,
      "language_loss": 0.76655769,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.78846371,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 2.6795809268951416
    },
    {
      "auxiliary_loss_clip": 0.01164557,
      "auxiliary_loss_mlp": 0.01026494,
      "balance_loss_clip": 1.01086354,
      "balance_loss_mlp": 1.01941609,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.7206884801982045,
      "language_loss": 0.79797971,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.81989014,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 2.7646076679229736
    },
    {
      "auxiliary_loss_clip": 0.01168955,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 0.93560439,
      "balance_loss_mlp": 1.02051425,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 2.2417896519041705,
      "language_loss": 0.72490978,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.74687862,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.850757122039795
    },
    {
      "auxiliary_loss_clip": 0.011722,
      "auxiliary_loss_mlp": 0.01027575,
      "balance_loss_clip": 0.93149173,
      "balance_loss_mlp": 1.01973712,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.7345087509705648,
      "language_loss": 0.64363599,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.6656338,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.885523557662964
    },
    {
      "auxiliary_loss_clip": 0.01156792,
      "auxiliary_loss_mlp": 0.01021524,
      "balance_loss_clip": 0.89500362,
      "balance_loss_mlp": 1.01492023,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 1.9552688936207596,
      "language_loss": 0.82334423,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.84512746,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.798861026763916
    },
    {
      "auxiliary_loss_clip": 0.01069863,
      "auxiliary_loss_mlp": 0.01001555,
      "balance_loss_clip": 0.90123272,
      "balance_loss_mlp": 1.00001705,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9528635368987325,
      "language_loss": 0.67382592,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69454014,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.209299087524414
    },
    {
      "auxiliary_loss_clip": 0.01171871,
      "auxiliary_loss_mlp": 0.01026554,
      "balance_loss_clip": 1.04893994,
      "balance_loss_mlp": 1.01874328,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 1.9097787641897401,
      "language_loss": 0.68586147,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.70784569,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.7186498641967773
    },
    {
      "auxiliary_loss_clip": 0.01153462,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 0.93115789,
      "balance_loss_mlp": 1.02132177,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 2.952035760921811,
      "language_loss": 0.82426786,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84608144,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 2.6761748790740967
    },
    {
      "auxiliary_loss_clip": 0.01166324,
      "auxiliary_loss_mlp": 0.010245,
      "balance_loss_clip": 0.97175539,
      "balance_loss_mlp": 1.01684642,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 2.0610224777608956,
      "language_loss": 0.75952768,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.78143585,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.7789249420166016
    },
    {
      "auxiliary_loss_clip": 0.01162113,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 1.00752366,
      "balance_loss_mlp": 1.02134168,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 1.8862901096097484,
      "language_loss": 0.83182871,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85373068,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 3.4175968170166016
    },
    {
      "auxiliary_loss_clip": 0.01066834,
      "auxiliary_loss_mlp": 0.01002066,
      "balance_loss_clip": 0.97321016,
      "balance_loss_mlp": 1.00048089,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.8761124220792527,
      "language_loss": 0.54680616,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56749517,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 3.293315887451172
    },
    {
      "auxiliary_loss_clip": 0.01154168,
      "auxiliary_loss_mlp": 0.01027223,
      "balance_loss_clip": 0.89328516,
      "balance_loss_mlp": 1.02042222,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 2.124042433471504,
      "language_loss": 0.71438384,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73619777,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.5717625617980957
    },
    {
      "auxiliary_loss_clip": 0.01158651,
      "auxiliary_loss_mlp": 0.01029338,
      "balance_loss_clip": 1.00808108,
      "balance_loss_mlp": 1.02167249,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 12.712136759664002,
      "language_loss": 0.6874432,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.70932305,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.5361087322235107
    },
    {
      "auxiliary_loss_clip": 0.01167254,
      "auxiliary_loss_mlp": 0.0102556,
      "balance_loss_clip": 1.00943649,
      "balance_loss_mlp": 1.01809144,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 2.377023779312224,
      "language_loss": 0.8305366,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.8524648,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 3.362377405166626
    },
    {
      "auxiliary_loss_clip": 0.01167025,
      "auxiliary_loss_mlp": 0.01026837,
      "balance_loss_clip": 1.04689217,
      "balance_loss_mlp": 1.01982749,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 1.961375145301958,
      "language_loss": 0.72229642,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74423504,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 3.818164348602295
    },
    {
      "auxiliary_loss_clip": 0.0116242,
      "auxiliary_loss_mlp": 0.01027167,
      "balance_loss_clip": 0.93441969,
      "balance_loss_mlp": 1.02010381,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.5629920687486925,
      "language_loss": 0.77334183,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.79523772,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 3.797006607055664
    },
    {
      "auxiliary_loss_clip": 0.01163721,
      "auxiliary_loss_mlp": 0.01022374,
      "balance_loss_clip": 0.92927134,
      "balance_loss_mlp": 1.01561522,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.4635684377088232,
      "language_loss": 0.68960321,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71146417,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.813478946685791
    },
    {
      "auxiliary_loss_clip": 0.0116834,
      "auxiliary_loss_mlp": 0.01027837,
      "balance_loss_clip": 1.01099205,
      "balance_loss_mlp": 1.02019012,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 3.1839700953956274,
      "language_loss": 0.73706305,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.75902474,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.61019229888916
    },
    {
      "auxiliary_loss_clip": 0.01161703,
      "auxiliary_loss_mlp": 0.01025802,
      "balance_loss_clip": 1.00901341,
      "balance_loss_mlp": 1.01907265,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 32.8506760739276,
      "language_loss": 0.87340355,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89527857,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.701042652130127
    },
    {
      "auxiliary_loss_clip": 0.01164338,
      "auxiliary_loss_mlp": 0.01020521,
      "balance_loss_clip": 1.04808283,
      "balance_loss_mlp": 1.01399708,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.4935016126811886,
      "language_loss": 0.73143971,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75328827,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 2.6508142948150635
    },
    {
      "auxiliary_loss_clip": 0.01164598,
      "auxiliary_loss_mlp": 0.0102639,
      "balance_loss_clip": 1.01007533,
      "balance_loss_mlp": 1.01986003,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 2.009750180893604,
      "language_loss": 0.63972306,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66163296,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 2.7136685848236084
    },
    {
      "auxiliary_loss_clip": 0.01166261,
      "auxiliary_loss_mlp": 0.01024768,
      "balance_loss_clip": 0.93476057,
      "balance_loss_mlp": 1.01810455,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 1.6564261848320587,
      "language_loss": 0.81543183,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.83734214,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 2.7571351528167725
    },
    {
      "auxiliary_loss_clip": 0.01164251,
      "auxiliary_loss_mlp": 0.01026358,
      "balance_loss_clip": 0.97006261,
      "balance_loss_mlp": 1.01928902,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 1.748321616967585,
      "language_loss": 0.74198151,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76388764,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.699960947036743
    },
    {
      "auxiliary_loss_clip": 0.01165124,
      "auxiliary_loss_mlp": 0.01025721,
      "balance_loss_clip": 0.96843863,
      "balance_loss_mlp": 1.01777577,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 2.097803172931682,
      "language_loss": 0.74599999,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.76790839,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 2.7183456420898438
    },
    {
      "auxiliary_loss_clip": 0.01163656,
      "auxiliary_loss_mlp": 0.01026391,
      "balance_loss_clip": 0.8931725,
      "balance_loss_mlp": 1.01898539,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 2.564190376633595,
      "language_loss": 0.77309632,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.79499674,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.7580528259277344
    },
    {
      "auxiliary_loss_clip": 0.01154023,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 0.93092924,
      "balance_loss_mlp": 1.01569247,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 2.171822485696211,
      "language_loss": 0.66398489,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68575561,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.7531731128692627
    },
    {
      "auxiliary_loss_clip": 0.0117311,
      "auxiliary_loss_mlp": 0.01028797,
      "balance_loss_clip": 1.01152039,
      "balance_loss_mlp": 1.02096462,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 2.256236199875772,
      "language_loss": 0.68558389,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.70760298,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.639767646789551
    },
    {
      "auxiliary_loss_clip": 0.01170761,
      "auxiliary_loss_mlp": 0.01026247,
      "balance_loss_clip": 0.85730475,
      "balance_loss_mlp": 1.01930606,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 1.7642411708773105,
      "language_loss": 0.66056383,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.68253392,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 2.8509438037872314
    },
    {
      "auxiliary_loss_clip": 0.0115424,
      "auxiliary_loss_mlp": 0.01027668,
      "balance_loss_clip": 0.93180591,
      "balance_loss_mlp": 1.02108455,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 1.7300286964358735,
      "language_loss": 0.82413983,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.84595895,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.7121660709381104
    },
    {
      "auxiliary_loss_clip": 0.01168227,
      "auxiliary_loss_mlp": 0.01029671,
      "balance_loss_clip": 0.97513664,
      "balance_loss_mlp": 1.02271533,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.1133581131812353,
      "language_loss": 0.78422064,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80619967,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.756455183029175
    },
    {
      "auxiliary_loss_clip": 0.0116099,
      "auxiliary_loss_mlp": 0.01121873,
      "balance_loss_clip": 1.00855494,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.5006660128799587,
      "language_loss": 0.83904743,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.86187607,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.6836225986480713
    },
    {
      "auxiliary_loss_clip": 0.01167578,
      "auxiliary_loss_mlp": 0.0102775,
      "balance_loss_clip": 1.01021266,
      "balance_loss_mlp": 1.02056134,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 1.7871443068558222,
      "language_loss": 0.73580384,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.75775719,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.68068528175354
    },
    {
      "auxiliary_loss_clip": 0.01067149,
      "auxiliary_loss_mlp": 0.01001098,
      "balance_loss_clip": 0.97429717,
      "balance_loss_mlp": 0.99952489,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8197101894690632,
      "language_loss": 0.54586512,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56654757,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.139392852783203
    },
    {
      "auxiliary_loss_clip": 0.01162307,
      "auxiliary_loss_mlp": 0.01020388,
      "balance_loss_clip": 0.97106695,
      "balance_loss_mlp": 1.01394749,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 1.620628677697643,
      "language_loss": 0.56044269,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58226955,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.6869664192199707
    },
    {
      "auxiliary_loss_clip": 0.01163281,
      "auxiliary_loss_mlp": 0.01022425,
      "balance_loss_clip": 0.93086326,
      "balance_loss_mlp": 1.01570749,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 2.0263100394162548,
      "language_loss": 0.75874233,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78059936,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.7482519149780273
    },
    {
      "auxiliary_loss_clip": 0.01164663,
      "auxiliary_loss_mlp": 0.01122299,
      "balance_loss_clip": 0.93063951,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 1.9805379571892927,
      "language_loss": 0.82065779,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.84352744,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 3.7325103282928467
    },
    {
      "auxiliary_loss_clip": 0.01170251,
      "auxiliary_loss_mlp": 0.01035648,
      "balance_loss_clip": 0.97122037,
      "balance_loss_mlp": 1.02835846,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 2.0469556637308273,
      "language_loss": 0.68242943,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70448852,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 2.692479133605957
    },
    {
      "auxiliary_loss_clip": 0.01164956,
      "auxiliary_loss_mlp": 0.0102754,
      "balance_loss_clip": 0.97788417,
      "balance_loss_mlp": 1.02075076,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 1.8570578551925665,
      "language_loss": 0.81820798,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.84013295,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 2.7327451705932617
    },
    {
      "auxiliary_loss_clip": 0.01071236,
      "auxiliary_loss_mlp": 0.01002038,
      "balance_loss_clip": 0.93577421,
      "balance_loss_mlp": 1.00047636,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.856473262750231,
      "language_loss": 0.65463841,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67537117,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.2015902996063232
    },
    {
      "auxiliary_loss_clip": 0.01161376,
      "auxiliary_loss_mlp": 0.01026683,
      "balance_loss_clip": 0.93594193,
      "balance_loss_mlp": 1.01976943,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 1.787230057644885,
      "language_loss": 0.74133706,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76321775,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 4.828469514846802
    },
    {
      "auxiliary_loss_clip": 0.0116675,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.00807345,
      "balance_loss_mlp": 1.02026999,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 2.347463105521853,
      "language_loss": 0.73864019,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76057887,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.7172930240631104
    },
    {
      "auxiliary_loss_clip": 0.01166408,
      "auxiliary_loss_mlp": 0.0112262,
      "balance_loss_clip": 0.9691726,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 1.7602363684525824,
      "language_loss": 0.81815016,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.84104049,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 3.629103660583496
    },
    {
      "auxiliary_loss_clip": 0.01160116,
      "auxiliary_loss_mlp": 0.01023639,
      "balance_loss_clip": 0.96638632,
      "balance_loss_mlp": 1.0166142,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 1.9087911798527954,
      "language_loss": 0.74582374,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.76766127,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.699247360229492
    },
    {
      "auxiliary_loss_clip": 0.01171128,
      "auxiliary_loss_mlp": 0.01027935,
      "balance_loss_clip": 1.00904882,
      "balance_loss_mlp": 1.02083325,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.2126959459728304,
      "language_loss": 0.74696946,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.76896012,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.673424243927002
    },
    {
      "auxiliary_loss_clip": 0.01166947,
      "auxiliary_loss_mlp": 0.01026135,
      "balance_loss_clip": 0.97257274,
      "balance_loss_mlp": 1.01937938,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 1.6397505897296867,
      "language_loss": 0.67704266,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.69897354,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.6646945476531982
    },
    {
      "auxiliary_loss_clip": 0.01154364,
      "auxiliary_loss_mlp": 0.01020847,
      "balance_loss_clip": 0.93088412,
      "balance_loss_mlp": 1.0139271,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.7605859651569598,
      "language_loss": 0.82274091,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84449297,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.7117371559143066
    },
    {
      "auxiliary_loss_clip": 0.01156834,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 0.89160383,
      "balance_loss_mlp": 1.02134585,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 2.3954307070507057,
      "language_loss": 0.65988886,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.68174124,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.742088556289673
    },
    {
      "auxiliary_loss_clip": 0.01167492,
      "auxiliary_loss_mlp": 0.01020859,
      "balance_loss_clip": 1.0465734,
      "balance_loss_mlp": 1.0133791,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 2.48250278948384,
      "language_loss": 0.61862516,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64050865,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 2.5855462551116943
    },
    {
      "auxiliary_loss_clip": 0.01168301,
      "auxiliary_loss_mlp": 0.0102336,
      "balance_loss_clip": 1.04677999,
      "balance_loss_mlp": 1.01573634,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 1.9308329445845653,
      "language_loss": 0.69413823,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71605486,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.6512553691864014
    },
    {
      "auxiliary_loss_clip": 0.01162476,
      "auxiliary_loss_mlp": 0.01023231,
      "balance_loss_clip": 0.9694792,
      "balance_loss_mlp": 1.01577103,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 2.6091634735703115,
      "language_loss": 0.69325805,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.71511507,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 2.6822242736816406
    },
    {
      "auxiliary_loss_clip": 0.01167622,
      "auxiliary_loss_mlp": 0.01025316,
      "balance_loss_clip": 0.89530921,
      "balance_loss_mlp": 1.01755536,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 1.7974205820491729,
      "language_loss": 0.84272206,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.86465144,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.738312244415283
    },
    {
      "auxiliary_loss_clip": 0.01165235,
      "auxiliary_loss_mlp": 0.01027375,
      "balance_loss_clip": 0.96943998,
      "balance_loss_mlp": 1.02042866,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 2.924910570823401,
      "language_loss": 0.8325792,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.8545053,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.709172487258911
    },
    {
      "auxiliary_loss_clip": 0.01166951,
      "auxiliary_loss_mlp": 0.01122406,
      "balance_loss_clip": 1.01268041,
      "balance_loss_mlp": 0.0,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 1.9066721480189366,
      "language_loss": 0.84199423,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.86488777,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 2.7296714782714844
    },
    {
      "auxiliary_loss_clip": 0.01157424,
      "auxiliary_loss_mlp": 0.01122335,
      "balance_loss_clip": 1.00678778,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 1.5977918396999367,
      "language_loss": 0.64067245,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.66347003,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 2.7142081260681152
    },
    {
      "auxiliary_loss_clip": 0.01173116,
      "auxiliary_loss_mlp": 0.01122236,
      "balance_loss_clip": 0.97450602,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 1.8317223684755528,
      "language_loss": 0.80121082,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.82416433,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.702597141265869
    },
    {
      "auxiliary_loss_clip": 0.01165829,
      "auxiliary_loss_mlp": 0.01022366,
      "balance_loss_clip": 0.97423095,
      "balance_loss_mlp": 1.01479352,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.5376364453853717,
      "language_loss": 0.76348388,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78536576,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 2.678269147872925
    },
    {
      "auxiliary_loss_clip": 0.01161268,
      "auxiliary_loss_mlp": 0.01023172,
      "balance_loss_clip": 0.97146189,
      "balance_loss_mlp": 1.01592398,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.4768018981390352,
      "language_loss": 0.85411417,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.87595856,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.6847870349884033
    },
    {
      "auxiliary_loss_clip": 0.01161806,
      "auxiliary_loss_mlp": 0.01023525,
      "balance_loss_clip": 0.97153682,
      "balance_loss_mlp": 1.01640224,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.3658065232633807,
      "language_loss": 0.7161392,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.73799247,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.6959228515625
    },
    {
      "auxiliary_loss_clip": 0.01163184,
      "auxiliary_loss_mlp": 0.0102438,
      "balance_loss_clip": 1.00747061,
      "balance_loss_mlp": 1.01780224,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 1.8726115414360138,
      "language_loss": 0.70439649,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72627211,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.647007703781128
    },
    {
      "auxiliary_loss_clip": 0.01166788,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 0.93275529,
      "balance_loss_mlp": 1.0178566,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 6.711575898052772,
      "language_loss": 0.57950294,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.60142094,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.757601022720337
    },
    {
      "auxiliary_loss_clip": 0.0116354,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 0.93552935,
      "balance_loss_mlp": 1.01965165,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.5742085204782619,
      "language_loss": 0.69945598,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72135651,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 2.810659408569336
    },
    {
      "auxiliary_loss_clip": 0.01166372,
      "auxiliary_loss_mlp": 0.01027288,
      "balance_loss_clip": 0.93672454,
      "balance_loss_mlp": 1.01958704,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 2.0703420995163686,
      "language_loss": 0.71494585,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73688245,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 3.535447120666504
    },
    {
      "auxiliary_loss_clip": 0.01164115,
      "auxiliary_loss_mlp": 0.01026755,
      "balance_loss_clip": 0.97217733,
      "balance_loss_mlp": 1.01988554,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 1.942780423773983,
      "language_loss": 0.77779138,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.79970002,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 2.7168731689453125
    },
    {
      "auxiliary_loss_clip": 0.01165655,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.00704682,
      "balance_loss_mlp": 1.0172689,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 4.184339659904399,
      "language_loss": 0.79286313,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81476438,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 2.616260051727295
    },
    {
      "auxiliary_loss_clip": 0.01170995,
      "auxiliary_loss_mlp": 0.01024237,
      "balance_loss_clip": 1.05009031,
      "balance_loss_mlp": 1.01715016,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 1.8532087109082713,
      "language_loss": 0.74885416,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.77080649,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.6063597202301025
    },
    {
      "auxiliary_loss_clip": 0.01161438,
      "auxiliary_loss_mlp": 0.0102873,
      "balance_loss_clip": 1.00993824,
      "balance_loss_mlp": 1.0221169,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 1.8566214055808288,
      "language_loss": 0.8512156,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.87311733,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 4.489595413208008
    },
    {
      "auxiliary_loss_clip": 0.01166699,
      "auxiliary_loss_mlp": 0.01122144,
      "balance_loss_clip": 1.00937784,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 1.8136679480100069,
      "language_loss": 0.83867449,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.86156291,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 2.6676385402679443
    },
    {
      "auxiliary_loss_clip": 0.01165275,
      "auxiliary_loss_mlp": 0.01026251,
      "balance_loss_clip": 0.97241658,
      "balance_loss_mlp": 1.01889825,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 2.1140338484308736,
      "language_loss": 0.82629329,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.84820855,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 3.6742310523986816
    },
    {
      "auxiliary_loss_clip": 0.01162862,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 0.97219783,
      "balance_loss_mlp": 1.02046454,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 2.2872894888511985,
      "language_loss": 0.71443266,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.73633683,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.7600085735321045
    },
    {
      "auxiliary_loss_clip": 0.01166143,
      "auxiliary_loss_mlp": 0.01024382,
      "balance_loss_clip": 1.01155066,
      "balance_loss_mlp": 1.01710975,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 1.5766310750347878,
      "language_loss": 0.77978581,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80169106,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.784630298614502
    },
    {
      "auxiliary_loss_clip": 0.01172827,
      "auxiliary_loss_mlp": 0.01029463,
      "balance_loss_clip": 1.04901052,
      "balance_loss_mlp": 1.02154183,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 1.9187304143781108,
      "language_loss": 0.72081387,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74283683,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.5744423866271973
    },
    {
      "auxiliary_loss_clip": 0.01161651,
      "auxiliary_loss_mlp": 0.01122776,
      "balance_loss_clip": 0.85539579,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 1.7287041249213297,
      "language_loss": 0.76407152,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.78691584,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.7410731315612793
    },
    {
      "auxiliary_loss_clip": 0.01169052,
      "auxiliary_loss_mlp": 0.01021758,
      "balance_loss_clip": 1.05005181,
      "balance_loss_mlp": 1.01536798,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 2.0788228477119284,
      "language_loss": 0.85658044,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.87848854,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 2.6987102031707764
    },
    {
      "auxiliary_loss_clip": 0.01170127,
      "auxiliary_loss_mlp": 0.010262,
      "balance_loss_clip": 1.00977039,
      "balance_loss_mlp": 1.01885033,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 2.095471168404814,
      "language_loss": 0.75225592,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77421916,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 2.690639019012451
    },
    {
      "auxiliary_loss_clip": 0.01165845,
      "auxiliary_loss_mlp": 0.01022545,
      "balance_loss_clip": 1.04703164,
      "balance_loss_mlp": 1.015553,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 2.23292453994803,
      "language_loss": 0.74155003,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76343393,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.65523624420166
    },
    {
      "auxiliary_loss_clip": 0.01167294,
      "auxiliary_loss_mlp": 0.01026409,
      "balance_loss_clip": 1.0080421,
      "balance_loss_mlp": 1.01943791,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 1.5012999815794386,
      "language_loss": 0.78463078,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.80656779,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.6803030967712402
    },
    {
      "auxiliary_loss_clip": 0.01161381,
      "auxiliary_loss_mlp": 0.01025229,
      "balance_loss_clip": 0.89640266,
      "balance_loss_mlp": 1.017349,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 1.7143278452751798,
      "language_loss": 0.70637852,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.7282446,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.7829792499542236
    },
    {
      "auxiliary_loss_clip": 0.01163925,
      "auxiliary_loss_mlp": 0.01122128,
      "balance_loss_clip": 1.0081811,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 1.9600887506761555,
      "language_loss": 0.65226406,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.67512453,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.683136224746704
    },
    {
      "auxiliary_loss_clip": 0.01166379,
      "auxiliary_loss_mlp": 0.01029316,
      "balance_loss_clip": 1.00854588,
      "balance_loss_mlp": 1.02206206,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 1.7365362390369363,
      "language_loss": 0.60453331,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62649024,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 2.6197686195373535
    },
    {
      "auxiliary_loss_clip": 0.01159622,
      "auxiliary_loss_mlp": 0.01023251,
      "balance_loss_clip": 1.00818181,
      "balance_loss_mlp": 1.01649785,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 2.410888369074166,
      "language_loss": 0.75504482,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.77687359,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 2.6027894020080566
    },
    {
      "auxiliary_loss_clip": 0.01175987,
      "auxiliary_loss_mlp": 0.01029963,
      "balance_loss_clip": 0.97713161,
      "balance_loss_mlp": 1.02201176,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.354896488946492,
      "language_loss": 0.74461496,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.76667452,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.6600725650787354
    },
    {
      "auxiliary_loss_clip": 0.01168679,
      "auxiliary_loss_mlp": 0.01020669,
      "balance_loss_clip": 0.97042745,
      "balance_loss_mlp": 1.01339388,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.5594937834755855,
      "language_loss": 0.7392202,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.7611137,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 2.7028796672821045
    },
    {
      "auxiliary_loss_clip": 0.01072074,
      "auxiliary_loss_mlp": 0.01002613,
      "balance_loss_clip": 0.90251535,
      "balance_loss_mlp": 1.00088429,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8843111768204993,
      "language_loss": 0.61803126,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63877809,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 3.4014196395874023
    },
    {
      "auxiliary_loss_clip": 0.01174771,
      "auxiliary_loss_mlp": 0.01024468,
      "balance_loss_clip": 1.00991392,
      "balance_loss_mlp": 1.0175004,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.5308904847884484,
      "language_loss": 0.91311967,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.93511212,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.6178364753723145
    },
    {
      "auxiliary_loss_clip": 0.01070886,
      "auxiliary_loss_mlp": 0.01004089,
      "balance_loss_clip": 0.93697178,
      "balance_loss_mlp": 1.00251567,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.7608266315359651,
      "language_loss": 0.59102547,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.61177522,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.364859104156494
    },
    {
      "auxiliary_loss_clip": 0.01152625,
      "auxiliary_loss_mlp": 0.0102369,
      "balance_loss_clip": 0.89384937,
      "balance_loss_mlp": 1.01664746,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 1.7276267152950329,
      "language_loss": 0.67739189,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69915509,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.8255536556243896
    },
    {
      "auxiliary_loss_clip": 0.01066751,
      "auxiliary_loss_mlp": 0.01004225,
      "balance_loss_clip": 0.97341168,
      "balance_loss_mlp": 1.00260341,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9347153019979894,
      "language_loss": 0.71472502,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.73543477,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.194619655609131
    },
    {
      "auxiliary_loss_clip": 0.0116705,
      "auxiliary_loss_mlp": 0.01023045,
      "balance_loss_clip": 1.01045096,
      "balance_loss_mlp": 1.01568961,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.417656064675074,
      "language_loss": 0.7302326,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.75213349,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 3.5092782974243164
    },
    {
      "auxiliary_loss_clip": 0.01165215,
      "auxiliary_loss_mlp": 0.01022784,
      "balance_loss_clip": 1.04811025,
      "balance_loss_mlp": 1.01613832,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 1.7972405206125586,
      "language_loss": 0.82052827,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.84240824,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.6271026134490967
    },
    {
      "auxiliary_loss_clip": 0.01167542,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 0.89662457,
      "balance_loss_mlp": 1.02076364,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.3725055273108753,
      "language_loss": 0.75710154,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77905983,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 2.7362277507781982
    },
    {
      "auxiliary_loss_clip": 0.01162213,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 0.97159237,
      "balance_loss_mlp": 1.02439761,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.06003849508506,
      "language_loss": 0.69999063,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.72193336,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 2.7106306552886963
    },
    {
      "auxiliary_loss_clip": 0.0106525,
      "auxiliary_loss_mlp": 0.01115733,
      "balance_loss_clip": 0.97562194,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7886209509160019,
      "language_loss": 0.55763149,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.57944131,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 5.1099629402160645
    },
    {
      "auxiliary_loss_clip": 0.01072663,
      "auxiliary_loss_mlp": 0.01001221,
      "balance_loss_clip": 0.94889683,
      "balance_loss_mlp": 0.99951631,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7864342202578544,
      "language_loss": 0.63745749,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65819633,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.1666619777679443
    },
    {
      "auxiliary_loss_clip": 0.01164712,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.01103187,
      "balance_loss_mlp": 1.01866829,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 3.4525680261426763,
      "language_loss": 0.77750444,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.79940611,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 3.618424892425537
    },
    {
      "auxiliary_loss_clip": 0.01169215,
      "auxiliary_loss_mlp": 0.01122059,
      "balance_loss_clip": 1.01159966,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 1.789974980350881,
      "language_loss": 0.68675935,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.70967209,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.674933433532715
    },
    {
      "auxiliary_loss_clip": 0.01169597,
      "auxiliary_loss_mlp": 0.01025435,
      "balance_loss_clip": 0.94063902,
      "balance_loss_mlp": 1.0188396,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 1.5746433872022618,
      "language_loss": 0.71460056,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73655087,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.7912862300872803
    },
    {
      "auxiliary_loss_clip": 0.01158996,
      "auxiliary_loss_mlp": 0.01028292,
      "balance_loss_clip": 0.96832848,
      "balance_loss_mlp": 1.02177453,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 1.8017653533595235,
      "language_loss": 0.85097808,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87285089,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.7051279544830322
    },
    {
      "auxiliary_loss_clip": 0.01160954,
      "auxiliary_loss_mlp": 0.0103014,
      "balance_loss_clip": 0.93685907,
      "balance_loss_mlp": 1.02290368,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 2.2777217390690576,
      "language_loss": 0.7394886,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.76139963,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.7034213542938232
    },
    {
      "auxiliary_loss_clip": 0.01161166,
      "auxiliary_loss_mlp": 0.01030299,
      "balance_loss_clip": 0.93319744,
      "balance_loss_mlp": 1.0229615,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 4.776309930939368,
      "language_loss": 0.6378175,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.6597321,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 2.692176342010498
    },
    {
      "auxiliary_loss_clip": 0.01162311,
      "auxiliary_loss_mlp": 0.01122737,
      "balance_loss_clip": 0.89658308,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 1.972031429754075,
      "language_loss": 0.76549065,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.78834116,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 2.736286163330078
    },
    {
      "auxiliary_loss_clip": 0.01164356,
      "auxiliary_loss_mlp": 0.01020119,
      "balance_loss_clip": 1.00875306,
      "balance_loss_mlp": 1.01352632,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 2.565219807043183,
      "language_loss": 0.76205719,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.78390193,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.669297218322754
    },
    {
      "auxiliary_loss_clip": 0.01164587,
      "auxiliary_loss_mlp": 0.01027095,
      "balance_loss_clip": 1.00849402,
      "balance_loss_mlp": 1.02025557,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 1.45029202075361,
      "language_loss": 0.75531733,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77723408,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.722832441329956
    },
    {
      "auxiliary_loss_clip": 0.01167804,
      "auxiliary_loss_mlp": 0.01028109,
      "balance_loss_clip": 1.04832935,
      "balance_loss_mlp": 1.02122474,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 1.8717850402654224,
      "language_loss": 0.7589916,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.78095067,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.6507341861724854
    },
    {
      "auxiliary_loss_clip": 0.01172185,
      "auxiliary_loss_mlp": 0.01031107,
      "balance_loss_clip": 0.96886635,
      "balance_loss_mlp": 1.02342677,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 2.7493340681736815,
      "language_loss": 0.72994411,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.75197703,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.6536848545074463
    },
    {
      "auxiliary_loss_clip": 0.01164351,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.04718733,
      "balance_loss_mlp": 1.01924562,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 2.1042877756838574,
      "language_loss": 0.75950921,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78141272,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 2.6180477142333984
    },
    {
      "auxiliary_loss_clip": 0.01165498,
      "auxiliary_loss_mlp": 0.01021235,
      "balance_loss_clip": 0.97210896,
      "balance_loss_mlp": 1.01414156,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 1.5996175596172086,
      "language_loss": 0.8390764,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.86094373,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.795624256134033
    },
    {
      "auxiliary_loss_clip": 0.01169337,
      "auxiliary_loss_mlp": 0.01021017,
      "balance_loss_clip": 1.01037669,
      "balance_loss_mlp": 1.01436543,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.9446880053841271,
      "language_loss": 0.75255799,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.77446151,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.6830506324768066
    },
    {
      "auxiliary_loss_clip": 0.01067987,
      "auxiliary_loss_mlp": 0.01001223,
      "balance_loss_clip": 0.97394979,
      "balance_loss_mlp": 0.99957764,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.9701609046699711,
      "language_loss": 0.58595675,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.6066488,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.234095811843872
    },
    {
      "auxiliary_loss_clip": 0.01166129,
      "auxiliary_loss_mlp": 0.01027979,
      "balance_loss_clip": 1.01129961,
      "balance_loss_mlp": 1.02142835,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 2.369423712376231,
      "language_loss": 0.77841121,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80035233,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 2.587348222732544
    },
    {
      "auxiliary_loss_clip": 0.01164479,
      "auxiliary_loss_mlp": 0.01121991,
      "balance_loss_clip": 0.89256859,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.613092114026478,
      "language_loss": 0.71385944,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.7367242,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.8248953819274902
    },
    {
      "auxiliary_loss_clip": 0.01159071,
      "auxiliary_loss_mlp": 0.01023094,
      "balance_loss_clip": 0.92776167,
      "balance_loss_mlp": 1.01598048,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 6.21826765184467,
      "language_loss": 0.82820964,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85003132,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.687966823577881
    },
    {
      "auxiliary_loss_clip": 0.01157907,
      "auxiliary_loss_mlp": 0.01024419,
      "balance_loss_clip": 1.00669813,
      "balance_loss_mlp": 1.01734662,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 2.2223588363444855,
      "language_loss": 0.7482751,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.77009833,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.609250545501709
    },
    {
      "auxiliary_loss_clip": 0.01077024,
      "auxiliary_loss_mlp": 0.01116713,
      "balance_loss_clip": 0.91174567,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8571628504954478,
      "language_loss": 0.65784937,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.67978674,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.277566432952881
    },
    {
      "auxiliary_loss_clip": 0.01168917,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 0.97253239,
      "balance_loss_mlp": 1.02141547,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 1.923612452563062,
      "language_loss": 0.73897165,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.76094389,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 3.367898941040039
    },
    {
      "auxiliary_loss_clip": 0.01166341,
      "auxiliary_loss_mlp": 0.01028568,
      "balance_loss_clip": 1.04552436,
      "balance_loss_mlp": 1.02104604,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 1.9358938102162082,
      "language_loss": 0.83948481,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86143392,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.6432831287384033
    },
    {
      "auxiliary_loss_clip": 0.01067859,
      "auxiliary_loss_mlp": 0.01001705,
      "balance_loss_clip": 0.93789601,
      "balance_loss_mlp": 1.00010777,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7676709602721373,
      "language_loss": 0.59092468,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61162031,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.2407758235931396
    },
    {
      "auxiliary_loss_clip": 0.01166735,
      "auxiliary_loss_mlp": 0.01024798,
      "balance_loss_clip": 1.04590511,
      "balance_loss_mlp": 1.01763391,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 2.3492687632647438,
      "language_loss": 0.73993623,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76185149,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 2.607318639755249
    },
    {
      "auxiliary_loss_clip": 0.01163759,
      "auxiliary_loss_mlp": 0.01024951,
      "balance_loss_clip": 0.89410633,
      "balance_loss_mlp": 1.01779866,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 2.5714318605776136,
      "language_loss": 0.7542218,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.77610886,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 3.6676547527313232
    },
    {
      "auxiliary_loss_clip": 0.01163896,
      "auxiliary_loss_mlp": 0.01026191,
      "balance_loss_clip": 0.93494892,
      "balance_loss_mlp": 1.01948202,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.652421696223589,
      "language_loss": 0.87705588,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.89895672,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.7024242877960205
    },
    {
      "auxiliary_loss_clip": 0.01159283,
      "auxiliary_loss_mlp": 0.01123012,
      "balance_loss_clip": 0.93059504,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 1.775180072994,
      "language_loss": 0.70918822,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.73201114,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 3.808515787124634
    },
    {
      "auxiliary_loss_clip": 0.01158017,
      "auxiliary_loss_mlp": 0.01023762,
      "balance_loss_clip": 0.93250591,
      "balance_loss_mlp": 1.0170114,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.8144624052602079,
      "language_loss": 0.69466418,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.71648192,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 2.9050776958465576
    },
    {
      "auxiliary_loss_clip": 0.01172852,
      "auxiliary_loss_mlp": 0.01028071,
      "balance_loss_clip": 1.05160546,
      "balance_loss_mlp": 1.02076626,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 2.811239645265645,
      "language_loss": 0.78758973,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.80959892,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.5776050090789795
    },
    {
      "auxiliary_loss_clip": 0.0116473,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.01197743,
      "balance_loss_mlp": 1.0208149,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 1.893381400909145,
      "language_loss": 0.77729136,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.79921877,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.641101121902466
    },
    {
      "auxiliary_loss_clip": 0.01163991,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.00918007,
      "balance_loss_mlp": 1.01872218,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 1.6715184205192455,
      "language_loss": 0.87197661,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89387178,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 2.701861619949341
    },
    {
      "auxiliary_loss_clip": 0.011717,
      "auxiliary_loss_mlp": 0.01028215,
      "balance_loss_clip": 1.04833198,
      "balance_loss_mlp": 1.02038574,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 2.8439497439966477,
      "language_loss": 0.64164877,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.66364789,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 2.5955357551574707
    },
    {
      "auxiliary_loss_clip": 0.01170669,
      "auxiliary_loss_mlp": 0.01028446,
      "balance_loss_clip": 0.93462199,
      "balance_loss_mlp": 1.02082229,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 1.7377336207097733,
      "language_loss": 0.83993804,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86192918,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 2.6394190788269043
    },
    {
      "auxiliary_loss_clip": 0.01158121,
      "auxiliary_loss_mlp": 0.01121897,
      "balance_loss_clip": 0.97028613,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 2.2726338530271897,
      "language_loss": 0.69140655,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.7142067,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.6650030612945557
    },
    {
      "auxiliary_loss_clip": 0.01168906,
      "auxiliary_loss_mlp": 0.0103009,
      "balance_loss_clip": 1.0126332,
      "balance_loss_mlp": 1.02253771,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 2.297009560042903,
      "language_loss": 0.75898409,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.78097403,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 2.6424689292907715
    },
    {
      "auxiliary_loss_clip": 0.01165418,
      "auxiliary_loss_mlp": 0.01022124,
      "balance_loss_clip": 1.04735923,
      "balance_loss_mlp": 1.01504862,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 1.6563527607760844,
      "language_loss": 0.74541461,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.76729006,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.654327869415283
    },
    {
      "auxiliary_loss_clip": 0.01176287,
      "auxiliary_loss_mlp": 0.01027146,
      "balance_loss_clip": 1.01141381,
      "balance_loss_mlp": 1.01998115,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 1.7574249181062396,
      "language_loss": 0.71752375,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.7395581,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.7054309844970703
    },
    {
      "auxiliary_loss_clip": 0.01163405,
      "auxiliary_loss_mlp": 0.01023319,
      "balance_loss_clip": 0.89679098,
      "balance_loss_mlp": 1.01702809,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 1.6575482279838731,
      "language_loss": 0.75310266,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77496994,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 2.8197832107543945
    },
    {
      "auxiliary_loss_clip": 0.01162688,
      "auxiliary_loss_mlp": 0.01027171,
      "balance_loss_clip": 1.00979352,
      "balance_loss_mlp": 1.01985431,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 1.9795894576058317,
      "language_loss": 0.82734382,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.84924239,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 2.639616012573242
    },
    {
      "auxiliary_loss_clip": 0.01163422,
      "auxiliary_loss_mlp": 0.01025426,
      "balance_loss_clip": 1.04611039,
      "balance_loss_mlp": 1.01809156,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 1.550166190886845,
      "language_loss": 0.64569801,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.66758645,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.6063129901885986
    },
    {
      "auxiliary_loss_clip": 0.01167841,
      "auxiliary_loss_mlp": 0.01021842,
      "balance_loss_clip": 1.04927254,
      "balance_loss_mlp": 1.01495194,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 4.141537924364451,
      "language_loss": 0.75786269,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77975953,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.6222329139709473
    },
    {
      "auxiliary_loss_clip": 0.01159672,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 0.81874913,
      "balance_loss_mlp": 1.02178669,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 2.1324336645682025,
      "language_loss": 0.78813255,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.8100152,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 2.7770752906799316
    },
    {
      "auxiliary_loss_clip": 0.01162023,
      "auxiliary_loss_mlp": 0.0102399,
      "balance_loss_clip": 0.89209402,
      "balance_loss_mlp": 1.01748419,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 2.0369162668341927,
      "language_loss": 0.65243661,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.6742968,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.8126046657562256
    },
    {
      "auxiliary_loss_clip": 0.0116349,
      "auxiliary_loss_mlp": 0.01032816,
      "balance_loss_clip": 0.96891832,
      "balance_loss_mlp": 1.02537155,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.3446800838111233,
      "language_loss": 0.79645824,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.8184213,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.7086167335510254
    },
    {
      "auxiliary_loss_clip": 0.01161936,
      "auxiliary_loss_mlp": 0.0102922,
      "balance_loss_clip": 1.00863361,
      "balance_loss_mlp": 1.02210307,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 1.5687694754535377,
      "language_loss": 0.80529213,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.82720363,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.6824207305908203
    },
    {
      "auxiliary_loss_clip": 0.0116638,
      "auxiliary_loss_mlp": 0.01028906,
      "balance_loss_clip": 1.04818273,
      "balance_loss_mlp": 1.02185512,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 2.4899223698824913,
      "language_loss": 0.72947431,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75142717,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 2.658154010772705
    },
    {
      "auxiliary_loss_clip": 0.01160407,
      "auxiliary_loss_mlp": 0.01026894,
      "balance_loss_clip": 0.89370024,
      "balance_loss_mlp": 1.01924372,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 2.006374905069691,
      "language_loss": 0.76042736,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78230041,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 3.617708921432495
    },
    {
      "auxiliary_loss_clip": 0.01163236,
      "auxiliary_loss_mlp": 0.01026414,
      "balance_loss_clip": 0.96994078,
      "balance_loss_mlp": 1.01928806,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.6509078476854464,
      "language_loss": 0.84186774,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86376429,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.7696480751037598
    },
    {
      "auxiliary_loss_clip": 0.01163914,
      "auxiliary_loss_mlp": 0.01026889,
      "balance_loss_clip": 1.00664926,
      "balance_loss_mlp": 1.02033603,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 2.0371660745269753,
      "language_loss": 0.81044555,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.83235359,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.7246975898742676
    },
    {
      "auxiliary_loss_clip": 0.01067115,
      "auxiliary_loss_mlp": 0.01002113,
      "balance_loss_clip": 0.97332478,
      "balance_loss_mlp": 1.00040841,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7680656777950574,
      "language_loss": 0.61355591,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.6342482,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.294503688812256
    },
    {
      "auxiliary_loss_clip": 0.01163143,
      "auxiliary_loss_mlp": 0.01026421,
      "balance_loss_clip": 1.00732899,
      "balance_loss_mlp": 1.01930118,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 1.5091937981208767,
      "language_loss": 0.64170974,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.66360539,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 4.5413923263549805
    },
    {
      "auxiliary_loss_clip": 0.01167807,
      "auxiliary_loss_mlp": 0.01027831,
      "balance_loss_clip": 0.93944502,
      "balance_loss_mlp": 1.02041936,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 2.0424899979425692,
      "language_loss": 0.72007871,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.74203509,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.730485200881958
    },
    {
      "auxiliary_loss_clip": 0.0116061,
      "auxiliary_loss_mlp": 0.0102815,
      "balance_loss_clip": 0.96967191,
      "balance_loss_mlp": 1.0202229,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 2.0842946703807868,
      "language_loss": 0.79321533,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.81510293,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 3.637333393096924
    },
    {
      "auxiliary_loss_clip": 0.01173481,
      "auxiliary_loss_mlp": 0.01024722,
      "balance_loss_clip": 0.93518519,
      "balance_loss_mlp": 1.0179956,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 2.4372750889143884,
      "language_loss": 0.7883271,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81030917,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.7304561138153076
    },
    {
      "auxiliary_loss_clip": 0.01162484,
      "auxiliary_loss_mlp": 0.01024247,
      "balance_loss_clip": 0.97396892,
      "balance_loss_mlp": 1.01763058,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.9219349898505715,
      "language_loss": 0.68203551,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.70390284,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.6578257083892822
    },
    {
      "auxiliary_loss_clip": 0.01161095,
      "auxiliary_loss_mlp": 0.01027006,
      "balance_loss_clip": 1.00799227,
      "balance_loss_mlp": 1.02010655,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 2.076901001376128,
      "language_loss": 0.7358669,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.75774789,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 2.711379289627075
    },
    {
      "auxiliary_loss_clip": 0.01158496,
      "auxiliary_loss_mlp": 0.01025514,
      "balance_loss_clip": 0.93068802,
      "balance_loss_mlp": 1.01858187,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 2.7828252245166456,
      "language_loss": 0.62491351,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.64675361,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 2.988229513168335
    },
    {
      "auxiliary_loss_clip": 0.01164974,
      "auxiliary_loss_mlp": 0.01027267,
      "balance_loss_clip": 1.00984526,
      "balance_loss_mlp": 1.01984656,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 1.706061282599936,
      "language_loss": 0.76927316,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79119551,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 2.8059864044189453
    },
    {
      "auxiliary_loss_clip": 0.01168409,
      "auxiliary_loss_mlp": 0.01025238,
      "balance_loss_clip": 1.04853296,
      "balance_loss_mlp": 1.01814234,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 1.664941980573841,
      "language_loss": 0.76091516,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.7828517,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.5981361865997314
    },
    {
      "auxiliary_loss_clip": 0.0116917,
      "auxiliary_loss_mlp": 0.01025692,
      "balance_loss_clip": 0.97360075,
      "balance_loss_mlp": 1.01868248,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 1.9696129434296714,
      "language_loss": 0.72601783,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.74796641,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.7443227767944336
    },
    {
      "auxiliary_loss_clip": 0.0116855,
      "auxiliary_loss_mlp": 0.01122349,
      "balance_loss_clip": 0.93290073,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 1.9371121897689967,
      "language_loss": 0.76618564,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.78909457,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 2.7693262100219727
    },
    {
      "auxiliary_loss_clip": 0.01167523,
      "auxiliary_loss_mlp": 0.01021815,
      "balance_loss_clip": 0.89340186,
      "balance_loss_mlp": 1.01524305,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.0745018943924203,
      "language_loss": 0.75114697,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.77304035,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.853425979614258
    },
    {
      "auxiliary_loss_clip": 0.01163649,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 0.97069693,
      "balance_loss_mlp": 1.02204382,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 3.1110095478433113,
      "language_loss": 0.60619164,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.62812787,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.737097978591919
    },
    {
      "auxiliary_loss_clip": 0.01161835,
      "auxiliary_loss_mlp": 0.01021289,
      "balance_loss_clip": 0.96871316,
      "balance_loss_mlp": 1.01373732,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 1.8143217906316793,
      "language_loss": 0.72639477,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.74822599,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 2.673717975616455
    },
    {
      "auxiliary_loss_clip": 0.01151721,
      "auxiliary_loss_mlp": 0.01025567,
      "balance_loss_clip": 0.89218426,
      "balance_loss_mlp": 1.0190438,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 2.314842937277097,
      "language_loss": 0.79123205,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81300497,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.7979423999786377
    },
    {
      "auxiliary_loss_clip": 0.01167653,
      "auxiliary_loss_mlp": 0.01028768,
      "balance_loss_clip": 1.00822949,
      "balance_loss_mlp": 1.02173424,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 1.5585092329573733,
      "language_loss": 0.75582534,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77778947,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.696974992752075
    },
    {
      "auxiliary_loss_clip": 0.01168904,
      "auxiliary_loss_mlp": 0.0102601,
      "balance_loss_clip": 0.97137749,
      "balance_loss_mlp": 1.01917946,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 1.9517951107238831,
      "language_loss": 0.76699781,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78894693,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.6465322971343994
    },
    {
      "auxiliary_loss_clip": 0.01160568,
      "auxiliary_loss_mlp": 0.01031432,
      "balance_loss_clip": 0.85468006,
      "balance_loss_mlp": 1.02423787,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 1.8687075501573778,
      "language_loss": 0.79257274,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.8144927,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 2.8572874069213867
    },
    {
      "auxiliary_loss_clip": 0.01158294,
      "auxiliary_loss_mlp": 0.01021951,
      "balance_loss_clip": 0.89399219,
      "balance_loss_mlp": 1.01443172,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 1.9259166161560086,
      "language_loss": 0.7669313,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.78873372,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.7706105709075928
    },
    {
      "auxiliary_loss_clip": 0.01169711,
      "auxiliary_loss_mlp": 0.01024533,
      "balance_loss_clip": 1.01091361,
      "balance_loss_mlp": 1.01642036,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 2.111821493619589,
      "language_loss": 0.69493908,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.71688145,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.6463592052459717
    },
    {
      "auxiliary_loss_clip": 0.01170551,
      "auxiliary_loss_mlp": 0.01025039,
      "balance_loss_clip": 1.05022931,
      "balance_loss_mlp": 1.01799083,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.0838366765712673,
      "language_loss": 0.77980363,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.80175948,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.6383895874023438
    },
    {
      "auxiliary_loss_clip": 0.01068359,
      "auxiliary_loss_mlp": 0.01001908,
      "balance_loss_clip": 0.89970112,
      "balance_loss_mlp": 1.00032222,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.7993506117842045,
      "language_loss": 0.54015666,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.56085932,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 3.328259229660034
    },
    {
      "auxiliary_loss_clip": 0.01161286,
      "auxiliary_loss_mlp": 0.01025471,
      "balance_loss_clip": 0.97078854,
      "balance_loss_mlp": 1.0181396,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 2.051208514224757,
      "language_loss": 0.68460125,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.70646882,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 3.6977059841156006
    },
    {
      "auxiliary_loss_clip": 0.01162156,
      "auxiliary_loss_mlp": 0.01031757,
      "balance_loss_clip": 0.97105718,
      "balance_loss_mlp": 1.02372789,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 2.1998990602495256,
      "language_loss": 0.68444157,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70638061,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.897272825241089
    },
    {
      "auxiliary_loss_clip": 0.01174689,
      "auxiliary_loss_mlp": 0.01025509,
      "balance_loss_clip": 1.01082802,
      "balance_loss_mlp": 1.0175401,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.118560755475185,
      "language_loss": 0.72117567,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.74317765,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.6967110633850098
    },
    {
      "auxiliary_loss_clip": 0.01160877,
      "auxiliary_loss_mlp": 0.01025086,
      "balance_loss_clip": 0.97108763,
      "balance_loss_mlp": 1.01776671,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 1.7897099228917768,
      "language_loss": 0.69878483,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.72064447,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 2.715240478515625
    },
    {
      "auxiliary_loss_clip": 0.01170667,
      "auxiliary_loss_mlp": 0.01029551,
      "balance_loss_clip": 0.93148565,
      "balance_loss_mlp": 1.02175426,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 1.7323419382470833,
      "language_loss": 0.75480402,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77680618,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 3.7294983863830566
    },
    {
      "auxiliary_loss_clip": 0.01174034,
      "auxiliary_loss_mlp": 0.01122625,
      "balance_loss_clip": 0.97422576,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 6.042672055956829,
      "language_loss": 0.77616549,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.79913211,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 3.644202947616577
    },
    {
      "auxiliary_loss_clip": 0.01166567,
      "auxiliary_loss_mlp": 0.01027891,
      "balance_loss_clip": 1.01252437,
      "balance_loss_mlp": 1.02094996,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 1.8461642139149685,
      "language_loss": 0.71001768,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73196232,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.667611837387085
    },
    {
      "auxiliary_loss_clip": 0.01148796,
      "auxiliary_loss_mlp": 0.01029326,
      "balance_loss_clip": 0.92940634,
      "balance_loss_mlp": 1.02205694,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 1.9905346009446552,
      "language_loss": 0.86286932,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.88465053,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.7120118141174316
    },
    {
      "auxiliary_loss_clip": 0.01168834,
      "auxiliary_loss_mlp": 0.01122687,
      "balance_loss_clip": 1.0097239,
      "balance_loss_mlp": 0.0,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 2.056133831739459,
      "language_loss": 0.81216919,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.83508432,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 2.688159942626953
    },
    {
      "auxiliary_loss_clip": 0.01168807,
      "auxiliary_loss_mlp": 0.01024691,
      "balance_loss_clip": 1.0488075,
      "balance_loss_mlp": 1.01703215,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 2.5904415820754503,
      "language_loss": 0.84925967,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87119466,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 2.592663049697876
    },
    {
      "auxiliary_loss_clip": 0.0116645,
      "auxiliary_loss_mlp": 0.01020202,
      "balance_loss_clip": 0.93501806,
      "balance_loss_mlp": 1.01365435,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 1.6837865215070953,
      "language_loss": 0.68433523,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70620179,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 2.804802417755127
    },
    {
      "auxiliary_loss_clip": 0.01160127,
      "auxiliary_loss_mlp": 0.01025482,
      "balance_loss_clip": 0.96990019,
      "balance_loss_mlp": 1.01847553,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 2.3219536669219956,
      "language_loss": 0.70544225,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.72729826,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 2.659514904022217
    },
    {
      "auxiliary_loss_clip": 0.01165138,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.00928545,
      "balance_loss_mlp": 1.02045286,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 1.6711382772082766,
      "language_loss": 0.85993457,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.881863,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.6284520626068115
    },
    {
      "auxiliary_loss_clip": 0.01162764,
      "auxiliary_loss_mlp": 0.01025189,
      "balance_loss_clip": 0.97205424,
      "balance_loss_mlp": 1.01855445,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 2.6379842836454594,
      "language_loss": 0.73526746,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75714695,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.8466813564300537
    },
    {
      "auxiliary_loss_clip": 0.01170005,
      "auxiliary_loss_mlp": 0.01023355,
      "balance_loss_clip": 0.97098076,
      "balance_loss_mlp": 1.01654816,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 2.1649711091054216,
      "language_loss": 0.78768194,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.80961549,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 2.666083335876465
    },
    {
      "auxiliary_loss_clip": 0.01175612,
      "auxiliary_loss_mlp": 0.01026651,
      "balance_loss_clip": 0.93581879,
      "balance_loss_mlp": 1.01965094,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 1.9987132724788281,
      "language_loss": 0.65245187,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.67447448,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.7505552768707275
    },
    {
      "auxiliary_loss_clip": 0.01167673,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 0.97408664,
      "balance_loss_mlp": 1.02067566,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 1.9157987233407947,
      "language_loss": 0.7175917,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.7395469,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.6765220165252686
    },
    {
      "auxiliary_loss_clip": 0.01166553,
      "auxiliary_loss_mlp": 0.01021502,
      "balance_loss_clip": 0.97356373,
      "balance_loss_mlp": 1.0142628,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 2.3985692965420005,
      "language_loss": 0.80851257,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.83039308,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 2.668623447418213
    },
    {
      "auxiliary_loss_clip": 0.01067308,
      "auxiliary_loss_mlp": 0.01002078,
      "balance_loss_clip": 0.97368777,
      "balance_loss_mlp": 1.00049257,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6699865725425013,
      "language_loss": 0.6043365,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.6250304,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.2846436500549316
    },
    {
      "auxiliary_loss_clip": 0.01156978,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 0.93008983,
      "balance_loss_mlp": 1.02139235,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.5878806639125402,
      "language_loss": 0.82856381,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.85041809,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.7551541328430176
    },
    {
      "auxiliary_loss_clip": 0.01153033,
      "auxiliary_loss_mlp": 0.0102984,
      "balance_loss_clip": 0.93157125,
      "balance_loss_mlp": 1.02257395,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 1.9825399942094628,
      "language_loss": 0.80109024,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82291895,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.6833760738372803
    },
    {
      "auxiliary_loss_clip": 0.01169878,
      "auxiliary_loss_mlp": 0.010236,
      "balance_loss_clip": 1.04754126,
      "balance_loss_mlp": 1.01600671,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 23.048967046997927,
      "language_loss": 0.6190511,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64098591,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.7041988372802734
    },
    {
      "auxiliary_loss_clip": 0.0117233,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.05061698,
      "balance_loss_mlp": 1.0181694,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 2.169005904424347,
      "language_loss": 0.85159463,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.87357217,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 2.623309373855591
    },
    {
      "auxiliary_loss_clip": 0.01170678,
      "auxiliary_loss_mlp": 0.01032585,
      "balance_loss_clip": 0.97657698,
      "balance_loss_mlp": 1.02517939,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 1.9471249410674045,
      "language_loss": 0.6398133,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.66184592,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.792675256729126
    },
    {
      "auxiliary_loss_clip": 0.0116565,
      "auxiliary_loss_mlp": 0.01024952,
      "balance_loss_clip": 0.97112411,
      "balance_loss_mlp": 1.01799321,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.6358747228493726,
      "language_loss": 0.71890926,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.74081534,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 2.7451555728912354
    },
    {
      "auxiliary_loss_clip": 0.01165531,
      "auxiliary_loss_mlp": 0.01029114,
      "balance_loss_clip": 1.01128983,
      "balance_loss_mlp": 1.02169299,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.513601850614429,
      "language_loss": 0.81406271,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83600914,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 3.4770474433898926
    },
    {
      "auxiliary_loss_clip": 0.01167958,
      "auxiliary_loss_mlp": 0.01020388,
      "balance_loss_clip": 0.97224236,
      "balance_loss_mlp": 1.01351893,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 2.2452184061482234,
      "language_loss": 0.79021436,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81209785,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.6371371746063232
    },
    {
      "auxiliary_loss_clip": 0.0116653,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.04862738,
      "balance_loss_mlp": 1.01887798,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 1.8493939180328842,
      "language_loss": 0.7951858,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.81711286,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.5854599475860596
    },
    {
      "auxiliary_loss_clip": 0.01166618,
      "auxiliary_loss_mlp": 0.01024701,
      "balance_loss_clip": 1.04680753,
      "balance_loss_mlp": 1.01773286,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.7066383528030817,
      "language_loss": 0.74891937,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77083254,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.5681397914886475
    },
    {
      "auxiliary_loss_clip": 0.01175631,
      "auxiliary_loss_mlp": 0.01026312,
      "balance_loss_clip": 0.97304344,
      "balance_loss_mlp": 1.0194788,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.6691269735905094,
      "language_loss": 0.83432281,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.85634226,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.6823177337646484
    },
    {
      "auxiliary_loss_clip": 0.01063137,
      "auxiliary_loss_mlp": 0.01003876,
      "balance_loss_clip": 1.01085138,
      "balance_loss_mlp": 1.00229073,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.9128533550415872,
      "language_loss": 0.59051824,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.61118841,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 5.059353351593018
    },
    {
      "auxiliary_loss_clip": 0.01163189,
      "auxiliary_loss_mlp": 0.01033948,
      "balance_loss_clip": 0.93188846,
      "balance_loss_mlp": 1.02607393,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 2.3426576920228848,
      "language_loss": 0.66262072,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.68459213,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.8166403770446777
    },
    {
      "auxiliary_loss_clip": 0.01168177,
      "auxiliary_loss_mlp": 0.0112248,
      "balance_loss_clip": 1.01446891,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.7083442253452765,
      "language_loss": 0.70629299,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.72919959,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 3.689237594604492
    },
    {
      "auxiliary_loss_clip": 0.0117296,
      "auxiliary_loss_mlp": 0.01024474,
      "balance_loss_clip": 1.01179063,
      "balance_loss_mlp": 1.01729774,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 1.5509772229276666,
      "language_loss": 0.70462787,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.7266022,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.8391501903533936
    },
    {
      "auxiliary_loss_clip": 0.01167986,
      "auxiliary_loss_mlp": 0.01025099,
      "balance_loss_clip": 0.93489885,
      "balance_loss_mlp": 1.01837862,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.0834964250013943,
      "language_loss": 0.73520315,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.75713396,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.7319891452789307
    },
    {
      "auxiliary_loss_clip": 0.01169382,
      "auxiliary_loss_mlp": 0.01028261,
      "balance_loss_clip": 0.89279288,
      "balance_loss_mlp": 1.02121282,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 3.9263814210376498,
      "language_loss": 0.61110312,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.63307953,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 2.7290165424346924
    },
    {
      "auxiliary_loss_clip": 0.01161312,
      "auxiliary_loss_mlp": 0.01026122,
      "balance_loss_clip": 0.97200644,
      "balance_loss_mlp": 1.01913595,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.117201321988177,
      "language_loss": 0.75357866,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.77545303,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 2.6097700595855713
    },
    {
      "auxiliary_loss_clip": 0.01173813,
      "auxiliary_loss_mlp": 0.01024607,
      "balance_loss_clip": 0.97253889,
      "balance_loss_mlp": 1.01766014,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 1.9202742005302949,
      "language_loss": 0.75120193,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.77318621,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.760012626647949
    },
    {
      "auxiliary_loss_clip": 0.01169519,
      "auxiliary_loss_mlp": 0.01028429,
      "balance_loss_clip": 1.01236236,
      "balance_loss_mlp": 1.02098489,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 1.4968864576376248,
      "language_loss": 0.69498718,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.71696663,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.6442952156066895
    },
    {
      "auxiliary_loss_clip": 0.01168993,
      "auxiliary_loss_mlp": 0.01020001,
      "balance_loss_clip": 1.0101757,
      "balance_loss_mlp": 1.01283681,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 1.729373645587011,
      "language_loss": 0.77368641,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.79557633,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.7386043071746826
    },
    {
      "auxiliary_loss_clip": 0.0116861,
      "auxiliary_loss_mlp": 0.01029104,
      "balance_loss_clip": 1.00933456,
      "balance_loss_mlp": 1.02184069,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 1.8144222111361823,
      "language_loss": 0.71872717,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.7407043,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 2.634258270263672
    },
    {
      "auxiliary_loss_clip": 0.01166951,
      "auxiliary_loss_mlp": 0.0102211,
      "balance_loss_clip": 1.01050067,
      "balance_loss_mlp": 1.01547289,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 1.8715466506140643,
      "language_loss": 0.73674095,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.75863159,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.655517578125
    },
    {
      "auxiliary_loss_clip": 0.01163996,
      "auxiliary_loss_mlp": 0.01027384,
      "balance_loss_clip": 1.00668907,
      "balance_loss_mlp": 1.02051783,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 1.6269144637800101,
      "language_loss": 0.68346643,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70538026,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 2.8426523208618164
    },
    {
      "auxiliary_loss_clip": 0.01170802,
      "auxiliary_loss_mlp": 0.01023249,
      "balance_loss_clip": 1.01090336,
      "balance_loss_mlp": 1.01577473,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.3869975107423111,
      "language_loss": 0.63432842,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.65626889,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.725027084350586
    },
    {
      "auxiliary_loss_clip": 0.01166805,
      "auxiliary_loss_mlp": 0.01025984,
      "balance_loss_clip": 1.04735112,
      "balance_loss_mlp": 1.01860523,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 2.2358871857164173,
      "language_loss": 0.77009964,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.79202759,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 2.615661859512329
    },
    {
      "auxiliary_loss_clip": 0.01063026,
      "auxiliary_loss_mlp": 0.01004269,
      "balance_loss_clip": 0.90003258,
      "balance_loss_mlp": 1.00270736,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7425770504794058,
      "language_loss": 0.58376503,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60443795,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.2331857681274414
    },
    {
      "auxiliary_loss_clip": 0.01067068,
      "auxiliary_loss_mlp": 0.01001431,
      "balance_loss_clip": 0.97451174,
      "balance_loss_mlp": 0.99982196,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8334262249357043,
      "language_loss": 0.56950748,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.59019244,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 3.3812825679779053
    },
    {
      "auxiliary_loss_clip": 0.01070208,
      "auxiliary_loss_mlp": 0.01002139,
      "balance_loss_clip": 0.93735218,
      "balance_loss_mlp": 1.00045824,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8785054242929401,
      "language_loss": 0.67474991,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69547337,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.150148630142212
    },
    {
      "auxiliary_loss_clip": 0.0116574,
      "auxiliary_loss_mlp": 0.01026351,
      "balance_loss_clip": 0.89865887,
      "balance_loss_mlp": 1.01910853,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.6759777074055857,
      "language_loss": 0.75964278,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78156364,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.75607967376709
    },
    {
      "auxiliary_loss_clip": 0.01166737,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 0.97327387,
      "balance_loss_mlp": 1.0202651,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 1.6171870421763277,
      "language_loss": 0.82982409,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85176927,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.659743309020996
    },
    {
      "auxiliary_loss_clip": 0.01170489,
      "auxiliary_loss_mlp": 0.0102466,
      "balance_loss_clip": 1.00954235,
      "balance_loss_mlp": 1.01806736,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.52750254133146,
      "language_loss": 0.75847471,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.78042614,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 2.7322983741760254
    },
    {
      "auxiliary_loss_clip": 0.01166694,
      "auxiliary_loss_mlp": 0.01027938,
      "balance_loss_clip": 0.89366901,
      "balance_loss_mlp": 1.02104163,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 2.0256366382631685,
      "language_loss": 0.85755682,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.87950313,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 3.6028270721435547
    },
    {
      "auxiliary_loss_clip": 0.01168482,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04772377,
      "balance_loss_mlp": 1.01873231,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 1.8173767371101195,
      "language_loss": 0.76321316,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78515661,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.567706823348999
    },
    {
      "auxiliary_loss_clip": 0.01165491,
      "auxiliary_loss_mlp": 0.01020763,
      "balance_loss_clip": 1.0080421,
      "balance_loss_mlp": 1.01322317,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.8897508221722343,
      "language_loss": 0.80242836,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82429087,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.657541036605835
    },
    {
      "auxiliary_loss_clip": 0.01166514,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 1.01008058,
      "balance_loss_mlp": 1.01979887,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 1.7403856939907842,
      "language_loss": 0.76450539,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.78643835,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.6254844665527344
    },
    {
      "auxiliary_loss_clip": 0.01168115,
      "auxiliary_loss_mlp": 0.01024717,
      "balance_loss_clip": 0.97026783,
      "balance_loss_mlp": 1.01733184,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.5858929034279727,
      "language_loss": 0.76955485,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79148316,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 3.712921142578125
    },
    {
      "auxiliary_loss_clip": 0.01162198,
      "auxiliary_loss_mlp": 0.0103154,
      "balance_loss_clip": 0.97186333,
      "balance_loss_mlp": 1.02415526,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 2.5281673903634263,
      "language_loss": 0.78462124,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.80655861,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.6823861598968506
    },
    {
      "auxiliary_loss_clip": 0.01169399,
      "auxiliary_loss_mlp": 0.01123037,
      "balance_loss_clip": 1.0093255,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 2.4645841063433087,
      "language_loss": 0.83214051,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.85506487,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 2.6647825241088867
    },
    {
      "auxiliary_loss_clip": 0.01167256,
      "auxiliary_loss_mlp": 0.01027693,
      "balance_loss_clip": 1.00852084,
      "balance_loss_mlp": 1.02092457,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 4.4597298823101275,
      "language_loss": 0.81876242,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84071195,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 3.6190550327301025
    },
    {
      "auxiliary_loss_clip": 0.01063559,
      "auxiliary_loss_mlp": 0.01000178,
      "balance_loss_clip": 1.01117706,
      "balance_loss_mlp": 0.9985804,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6853268323935653,
      "language_loss": 0.56026375,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.58090115,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.195819139480591
    },
    {
      "auxiliary_loss_clip": 0.01175001,
      "auxiliary_loss_mlp": 0.01035506,
      "balance_loss_clip": 0.97585511,
      "balance_loss_mlp": 1.02782273,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 1.7446409211884957,
      "language_loss": 0.76470298,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.78680801,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.882431745529175
    },
    {
      "auxiliary_loss_clip": 0.01170074,
      "auxiliary_loss_mlp": 0.01029698,
      "balance_loss_clip": 0.97336954,
      "balance_loss_mlp": 1.02226508,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 1.9164518167824611,
      "language_loss": 0.66704929,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.68904698,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 2.6748838424682617
    },
    {
      "auxiliary_loss_clip": 0.01168456,
      "auxiliary_loss_mlp": 0.01019501,
      "balance_loss_clip": 0.97287369,
      "balance_loss_mlp": 1.0122745,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 1.875961745516027,
      "language_loss": 0.76621413,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.78809369,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 2.7093865871429443
    },
    {
      "auxiliary_loss_clip": 0.01169748,
      "auxiliary_loss_mlp": 0.01024784,
      "balance_loss_clip": 1.01249218,
      "balance_loss_mlp": 1.01780438,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 4.736358257934801,
      "language_loss": 0.77654493,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.79849017,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.680497884750366
    },
    {
      "auxiliary_loss_clip": 0.01166664,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 0.97374129,
      "balance_loss_mlp": 1.0207808,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.5433337032190435,
      "language_loss": 0.79472256,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81666768,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.631200075149536
    },
    {
      "auxiliary_loss_clip": 0.01167883,
      "auxiliary_loss_mlp": 0.01030044,
      "balance_loss_clip": 1.05001163,
      "balance_loss_mlp": 1.02324271,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 1.6528059325843887,
      "language_loss": 0.74069858,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.76267785,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.621912956237793
    },
    {
      "auxiliary_loss_clip": 0.01157839,
      "auxiliary_loss_mlp": 0.01122825,
      "balance_loss_clip": 0.89284283,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 1.7014889047120425,
      "language_loss": 0.74317509,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.76598167,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 2.756920337677002
    },
    {
      "auxiliary_loss_clip": 0.01165589,
      "auxiliary_loss_mlp": 0.01024596,
      "balance_loss_clip": 0.97464669,
      "balance_loss_mlp": 1.01703548,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.468713498361715,
      "language_loss": 0.72102755,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.74292934,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.6464109420776367
    },
    {
      "auxiliary_loss_clip": 0.01166993,
      "auxiliary_loss_mlp": 0.01027492,
      "balance_loss_clip": 1.04703617,
      "balance_loss_mlp": 1.02008295,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 2.140152260603949,
      "language_loss": 0.727521,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74946582,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 2.617608070373535
    },
    {
      "auxiliary_loss_clip": 0.01164819,
      "auxiliary_loss_mlp": 0.01021928,
      "balance_loss_clip": 0.9690817,
      "balance_loss_mlp": 1.01466179,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 2.9839987472821488,
      "language_loss": 0.84888971,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.87075722,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.63505482673645
    },
    {
      "auxiliary_loss_clip": 0.01173034,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 0.93461752,
      "balance_loss_mlp": 1.02313352,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 1.8141353388549655,
      "language_loss": 0.66162854,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.68365914,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 2.6926522254943848
    },
    {
      "auxiliary_loss_clip": 0.01059739,
      "auxiliary_loss_mlp": 0.01001219,
      "balance_loss_clip": 0.93654859,
      "balance_loss_mlp": 0.99966925,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8437080040120781,
      "language_loss": 0.52634478,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54695433,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.227926015853882
    },
    {
      "auxiliary_loss_clip": 0.01156109,
      "auxiliary_loss_mlp": 0.01024093,
      "balance_loss_clip": 0.93138379,
      "balance_loss_mlp": 1.01661229,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 1.492684864648409,
      "language_loss": 0.73418248,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.75598449,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 2.7281696796417236
    },
    {
      "auxiliary_loss_clip": 0.01160506,
      "auxiliary_loss_mlp": 0.01025611,
      "balance_loss_clip": 0.97062075,
      "balance_loss_mlp": 1.01855421,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.837704408860677,
      "language_loss": 0.72045934,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74232054,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.663090467453003
    },
    {
      "auxiliary_loss_clip": 0.01163222,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 0.9326368,
      "balance_loss_mlp": 1.01790595,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 3.2213214986631065,
      "language_loss": 0.73141003,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75329691,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.7505998611450195
    },
    {
      "auxiliary_loss_clip": 0.01165757,
      "auxiliary_loss_mlp": 0.0102931,
      "balance_loss_clip": 1.00790191,
      "balance_loss_mlp": 1.02191305,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 2.0468413539173618,
      "language_loss": 0.78028506,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80223572,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 2.6446640491485596
    },
    {
      "auxiliary_loss_clip": 0.01153665,
      "auxiliary_loss_mlp": 0.01023871,
      "balance_loss_clip": 0.92729783,
      "balance_loss_mlp": 1.01700807,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.5704600697756521,
      "language_loss": 0.74561125,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.76738662,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 2.7883107662200928
    },
    {
      "auxiliary_loss_clip": 0.01170564,
      "auxiliary_loss_mlp": 0.01122956,
      "balance_loss_clip": 0.97181875,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 2.0019814840909165,
      "language_loss": 0.65980577,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.68274093,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 3.6363368034362793
    },
    {
      "auxiliary_loss_clip": 0.01062869,
      "auxiliary_loss_mlp": 0.01115883,
      "balance_loss_clip": 1.01068151,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.7205507838329649,
      "language_loss": 0.55791086,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.57969838,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.168738603591919
    },
    {
      "auxiliary_loss_clip": 0.01165679,
      "auxiliary_loss_mlp": 0.01031591,
      "balance_loss_clip": 1.00551319,
      "balance_loss_mlp": 1.02433133,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 2.2457711002339296,
      "language_loss": 0.83017588,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.85214853,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.5912086963653564
    },
    {
      "auxiliary_loss_clip": 0.0117108,
      "auxiliary_loss_mlp": 0.01025952,
      "balance_loss_clip": 1.04842019,
      "balance_loss_mlp": 1.01825452,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 3.4201516171631687,
      "language_loss": 0.83969522,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.86166555,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.5602259635925293
    },
    {
      "auxiliary_loss_clip": 0.01166611,
      "auxiliary_loss_mlp": 0.01024983,
      "balance_loss_clip": 1.00926661,
      "balance_loss_mlp": 1.01796472,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 4.9894547281845085,
      "language_loss": 0.74777883,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.76969481,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 3.4705746173858643
    },
    {
      "auxiliary_loss_clip": 0.01163982,
      "auxiliary_loss_mlp": 0.01026084,
      "balance_loss_clip": 0.97289711,
      "balance_loss_mlp": 1.01823378,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 2.7225467072069063,
      "language_loss": 0.81339598,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83529663,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 3.511343240737915
    },
    {
      "auxiliary_loss_clip": 0.01067122,
      "auxiliary_loss_mlp": 0.01002752,
      "balance_loss_clip": 0.97422671,
      "balance_loss_mlp": 1.0011189,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7170297600729217,
      "language_loss": 0.56323862,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.58393735,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.209611654281616
    },
    {
      "auxiliary_loss_clip": 0.0116079,
      "auxiliary_loss_mlp": 0.01026708,
      "balance_loss_clip": 0.9310447,
      "balance_loss_mlp": 1.01929605,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 1.9351719491497859,
      "language_loss": 0.74454331,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76641822,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 3.5997562408447266
    },
    {
      "auxiliary_loss_clip": 0.01167894,
      "auxiliary_loss_mlp": 0.01027367,
      "balance_loss_clip": 0.93379319,
      "balance_loss_mlp": 1.02049506,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.6630961596565128,
      "language_loss": 0.81896895,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.84092152,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 2.681600332260132
    },
    {
      "auxiliary_loss_clip": 0.01166306,
      "auxiliary_loss_mlp": 0.01029509,
      "balance_loss_clip": 1.01072955,
      "balance_loss_mlp": 1.02258873,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 1.7394988516087966,
      "language_loss": 0.680089,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.70204711,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.797266960144043
    },
    {
      "auxiliary_loss_clip": 0.01171668,
      "auxiliary_loss_mlp": 0.01029142,
      "balance_loss_clip": 0.97383642,
      "balance_loss_mlp": 1.02161062,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 3.415807245712423,
      "language_loss": 0.73650444,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.75851256,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 2.683889865875244
    },
    {
      "auxiliary_loss_clip": 0.01151927,
      "auxiliary_loss_mlp": 0.01026989,
      "balance_loss_clip": 0.85313904,
      "balance_loss_mlp": 1.01980042,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.1203321391723726,
      "language_loss": 0.81796646,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.83975565,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 2.754768133163452
    },
    {
      "auxiliary_loss_clip": 0.0115727,
      "auxiliary_loss_mlp": 0.01024655,
      "balance_loss_clip": 1.00625205,
      "balance_loss_mlp": 1.01752377,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 3.0060174452759494,
      "language_loss": 0.77383792,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79565716,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 2.7347018718719482
    },
    {
      "auxiliary_loss_clip": 0.01167518,
      "auxiliary_loss_mlp": 0.01025944,
      "balance_loss_clip": 0.9725793,
      "balance_loss_mlp": 1.01902735,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 1.932354136533387,
      "language_loss": 0.78932297,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.8112576,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.6270720958709717
    },
    {
      "auxiliary_loss_clip": 0.01169298,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 0.93479812,
      "balance_loss_mlp": 1.02016091,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 1.5344604406080777,
      "language_loss": 0.8322978,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.85427165,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.669914484024048
    },
    {
      "auxiliary_loss_clip": 0.01165763,
      "auxiliary_loss_mlp": 0.01031002,
      "balance_loss_clip": 0.97042567,
      "balance_loss_mlp": 1.0237509,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 1.8848464007767622,
      "language_loss": 0.80895531,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.83092296,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.6672046184539795
    },
    {
      "auxiliary_loss_clip": 0.01164096,
      "auxiliary_loss_mlp": 0.01022178,
      "balance_loss_clip": 1.00974154,
      "balance_loss_mlp": 1.01523387,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 2.0595409564116314,
      "language_loss": 0.76296794,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78483069,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 2.614588499069214
    },
    {
      "auxiliary_loss_clip": 0.01166711,
      "auxiliary_loss_mlp": 0.01023218,
      "balance_loss_clip": 1.01171768,
      "balance_loss_mlp": 1.01593447,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 1.7846939786879765,
      "language_loss": 0.84549814,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86739743,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 2.6299800872802734
    },
    {
      "auxiliary_loss_clip": 0.01165373,
      "auxiliary_loss_mlp": 0.01028636,
      "balance_loss_clip": 0.97492558,
      "balance_loss_mlp": 1.02188921,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.8763596374425757,
      "language_loss": 0.76023293,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.78217304,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.700995922088623
    },
    {
      "auxiliary_loss_clip": 0.01167413,
      "auxiliary_loss_mlp": 0.01021736,
      "balance_loss_clip": 0.89411378,
      "balance_loss_mlp": 1.01475644,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 1.8696132135115888,
      "language_loss": 0.71092904,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.73282051,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.694211959838867
    },
    {
      "auxiliary_loss_clip": 0.01168176,
      "auxiliary_loss_mlp": 0.01027501,
      "balance_loss_clip": 0.93438351,
      "balance_loss_mlp": 1.02044344,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 1.8839008243692594,
      "language_loss": 0.70191216,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72386891,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.7111666202545166
    },
    {
      "auxiliary_loss_clip": 0.01159861,
      "auxiliary_loss_mlp": 0.01026362,
      "balance_loss_clip": 0.97012967,
      "balance_loss_mlp": 1.01889384,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 1.980545554206769,
      "language_loss": 0.68461263,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.7064749,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 2.647355318069458
    },
    {
      "auxiliary_loss_clip": 0.011707,
      "auxiliary_loss_mlp": 0.0102857,
      "balance_loss_clip": 1.01065981,
      "balance_loss_mlp": 1.02088737,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 2.654847172004722,
      "language_loss": 0.64908683,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67107952,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.6315083503723145
    },
    {
      "auxiliary_loss_clip": 0.01063668,
      "auxiliary_loss_mlp": 0.01001415,
      "balance_loss_clip": 0.97446978,
      "balance_loss_mlp": 0.99981767,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.7958560994952903,
      "language_loss": 0.61450505,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63515592,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.20462965965271
    },
    {
      "auxiliary_loss_clip": 0.01172317,
      "auxiliary_loss_mlp": 0.01029431,
      "balance_loss_clip": 1.04889596,
      "balance_loss_mlp": 1.02165246,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.0367489277274538,
      "language_loss": 0.84144241,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86345989,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.628105640411377
    },
    {
      "auxiliary_loss_clip": 0.01166515,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.00929773,
      "balance_loss_mlp": 1.0187676,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 2.177292129338143,
      "language_loss": 0.80490577,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82682884,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 2.673638343811035
    },
    {
      "auxiliary_loss_clip": 0.01064865,
      "auxiliary_loss_mlp": 0.01001332,
      "balance_loss_clip": 0.97357696,
      "balance_loss_mlp": 0.99976999,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7377237117200126,
      "language_loss": 0.54994142,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.57060343,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 4.030500650405884
    },
    {
      "auxiliary_loss_clip": 0.01168947,
      "auxiliary_loss_mlp": 0.01020667,
      "balance_loss_clip": 1.01008821,
      "balance_loss_mlp": 1.01375318,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 1.8521307260486977,
      "language_loss": 0.76935965,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.79125583,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.604717493057251
    },
    {
      "auxiliary_loss_clip": 0.0116457,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 0.89351547,
      "balance_loss_mlp": 1.01906085,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.6312257295202097,
      "language_loss": 0.71756423,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.7394712,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.762772560119629
    },
    {
      "auxiliary_loss_clip": 0.01167537,
      "auxiliary_loss_mlp": 0.01028524,
      "balance_loss_clip": 1.0491581,
      "balance_loss_mlp": 1.0213412,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.7729744010009862,
      "language_loss": 0.87258315,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89454377,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.5636658668518066
    },
    {
      "auxiliary_loss_clip": 0.01161596,
      "auxiliary_loss_mlp": 0.01027081,
      "balance_loss_clip": 0.93365884,
      "balance_loss_mlp": 1.0202446,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 1.8455507685401986,
      "language_loss": 0.75082731,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77271414,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.7096426486968994
    },
    {
      "auxiliary_loss_clip": 0.01063432,
      "auxiliary_loss_mlp": 0.01001881,
      "balance_loss_clip": 0.97411048,
      "balance_loss_mlp": 1.00034285,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.8580538256086414,
      "language_loss": 0.61049145,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.63114458,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 5.053264379501343
    },
    {
      "auxiliary_loss_clip": 0.01167573,
      "auxiliary_loss_mlp": 0.0102897,
      "balance_loss_clip": 1.00946903,
      "balance_loss_mlp": 1.02165079,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.4864392162336477,
      "language_loss": 0.73514283,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75710827,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.6090571880340576
    },
    {
      "auxiliary_loss_clip": 0.01154968,
      "auxiliary_loss_mlp": 0.01024979,
      "balance_loss_clip": 0.9284839,
      "balance_loss_mlp": 1.01761222,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.5884377983454527,
      "language_loss": 0.59663969,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.6184392,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 3.6616551876068115
    },
    {
      "auxiliary_loss_clip": 0.01166543,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.01132095,
      "balance_loss_mlp": 1.02079773,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.4206864085310085,
      "language_loss": 0.89291912,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91486788,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 2.6465742588043213
    },
    {
      "auxiliary_loss_clip": 0.011694,
      "auxiliary_loss_mlp": 0.01030237,
      "balance_loss_clip": 1.01121545,
      "balance_loss_mlp": 1.02297688,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 1.7825656392486395,
      "language_loss": 0.81204081,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83403713,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 2.6330630779266357
    },
    {
      "auxiliary_loss_clip": 0.01168527,
      "auxiliary_loss_mlp": 0.0112285,
      "balance_loss_clip": 1.01277065,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 2.1054141436386518,
      "language_loss": 0.6910423,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.71395612,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 2.5620663166046143
    },
    {
      "auxiliary_loss_clip": 0.01154546,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 0.97031862,
      "balance_loss_mlp": 1.02314758,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.6649523598318046,
      "language_loss": 0.73768955,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.75953102,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 2.6027300357818604
    },
    {
      "auxiliary_loss_clip": 0.01166198,
      "auxiliary_loss_mlp": 0.01025674,
      "balance_loss_clip": 0.97131836,
      "balance_loss_mlp": 1.01864386,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 1.8109207755405394,
      "language_loss": 0.74947166,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.77139032,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 2.655012369155884
    },
    {
      "auxiliary_loss_clip": 0.01168511,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 0.97088742,
      "balance_loss_mlp": 1.01987815,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 1.9625963465251397,
      "language_loss": 0.81825292,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.84020984,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 2.708463668823242
    },
    {
      "auxiliary_loss_clip": 0.0115903,
      "auxiliary_loss_mlp": 0.01122254,
      "balance_loss_clip": 1.0081718,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 1.8440640681330853,
      "language_loss": 0.79153305,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.8143459,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.657048463821411
    },
    {
      "auxiliary_loss_clip": 0.01163903,
      "auxiliary_loss_mlp": 0.01028564,
      "balance_loss_clip": 1.00855398,
      "balance_loss_mlp": 1.02114964,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 2.3184072280818304,
      "language_loss": 0.73527873,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75720346,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.6516385078430176
    },
    {
      "auxiliary_loss_clip": 0.01170273,
      "auxiliary_loss_mlp": 0.01029366,
      "balance_loss_clip": 0.93638623,
      "balance_loss_mlp": 1.02195072,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 2.0112925062551392,
      "language_loss": 0.86789024,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.88988662,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 2.684767723083496
    },
    {
      "auxiliary_loss_clip": 0.01163482,
      "auxiliary_loss_mlp": 0.01026935,
      "balance_loss_clip": 0.93323761,
      "balance_loss_mlp": 1.02013707,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 3.8664016550164604,
      "language_loss": 0.79385853,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81576276,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 2.6788833141326904
    },
    {
      "auxiliary_loss_clip": 0.01166192,
      "auxiliary_loss_mlp": 0.01022642,
      "balance_loss_clip": 0.9737345,
      "balance_loss_mlp": 1.01599956,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.0098980384913276,
      "language_loss": 0.76142603,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.78331441,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 2.719014883041382
    },
    {
      "auxiliary_loss_clip": 0.01172778,
      "auxiliary_loss_mlp": 0.01025693,
      "balance_loss_clip": 0.97381032,
      "balance_loss_mlp": 1.01806986,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 2.25377912667196,
      "language_loss": 0.73753959,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.75952435,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.666167736053467
    },
    {
      "auxiliary_loss_clip": 0.01163239,
      "auxiliary_loss_mlp": 0.01024705,
      "balance_loss_clip": 0.89471811,
      "balance_loss_mlp": 1.01809514,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 2.2555624975429653,
      "language_loss": 0.90401566,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92589515,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.7639307975769043
    },
    {
      "auxiliary_loss_clip": 0.01158239,
      "auxiliary_loss_mlp": 0.01030286,
      "balance_loss_clip": 0.89567554,
      "balance_loss_mlp": 1.0223949,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 2.105992917466931,
      "language_loss": 0.77319098,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.79507625,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 2.677001953125
    },
    {
      "auxiliary_loss_clip": 0.01154704,
      "auxiliary_loss_mlp": 0.01025336,
      "balance_loss_clip": 0.97020304,
      "balance_loss_mlp": 1.01847851,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 1.8146602846041604,
      "language_loss": 0.78242254,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.80422294,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.6624624729156494
    },
    {
      "auxiliary_loss_clip": 0.01168601,
      "auxiliary_loss_mlp": 0.0102656,
      "balance_loss_clip": 1.01048195,
      "balance_loss_mlp": 1.01924682,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 1.7799998250650668,
      "language_loss": 0.73160994,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75356162,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.6181328296661377
    },
    {
      "auxiliary_loss_clip": 0.01168527,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.00944018,
      "balance_loss_mlp": 1.02057672,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 1.953716793149751,
      "language_loss": 0.71995783,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.7419132,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.8283326625823975
    },
    {
      "auxiliary_loss_clip": 0.01153684,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.00714111,
      "balance_loss_mlp": 1.02029169,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 2.1033111575265346,
      "language_loss": 0.8218888,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84369665,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 2.5936849117279053
    },
    {
      "auxiliary_loss_clip": 0.01167184,
      "auxiliary_loss_mlp": 0.0102863,
      "balance_loss_clip": 0.97166729,
      "balance_loss_mlp": 1.02146888,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.2011970021965928,
      "language_loss": 0.69840956,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.72036773,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.660841703414917
    },
    {
      "auxiliary_loss_clip": 0.01171813,
      "auxiliary_loss_mlp": 0.01023634,
      "balance_loss_clip": 1.04927504,
      "balance_loss_mlp": 1.01571274,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 3.5390868550958667,
      "language_loss": 0.81231254,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83426702,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 3.4251599311828613
    },
    {
      "auxiliary_loss_clip": 0.01162543,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 0.93438357,
      "balance_loss_mlp": 1.01961839,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.031678763225563,
      "language_loss": 0.87346172,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89535517,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.6526763439178467
    },
    {
      "auxiliary_loss_clip": 0.01176174,
      "auxiliary_loss_mlp": 0.01028879,
      "balance_loss_clip": 0.93646115,
      "balance_loss_mlp": 1.02134442,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 2.13374999493199,
      "language_loss": 0.71825147,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.74030203,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.7288169860839844
    },
    {
      "auxiliary_loss_clip": 0.01154074,
      "auxiliary_loss_mlp": 0.01023978,
      "balance_loss_clip": 0.89476573,
      "balance_loss_mlp": 1.01700974,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.5700970114577464,
      "language_loss": 0.77726394,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.79904449,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 4.533766269683838
    },
    {
      "auxiliary_loss_clip": 0.01159076,
      "auxiliary_loss_mlp": 0.01028807,
      "balance_loss_clip": 0.9346531,
      "balance_loss_mlp": 1.02185392,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 2.1346410553157935,
      "language_loss": 0.84722656,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.8691054,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.655207872390747
    },
    {
      "auxiliary_loss_clip": 0.01066102,
      "auxiliary_loss_mlp": 0.00999914,
      "balance_loss_clip": 0.89729548,
      "balance_loss_mlp": 0.99830478,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.7620758724970804,
      "language_loss": 0.58405292,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60471308,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.3585045337677
    },
    {
      "auxiliary_loss_clip": 0.01162201,
      "auxiliary_loss_mlp": 0.01029743,
      "balance_loss_clip": 0.97049499,
      "balance_loss_mlp": 1.02269769,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 1.903867621075791,
      "language_loss": 0.66050541,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68242478,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 3.7023766040802
    },
    {
      "auxiliary_loss_clip": 0.01167011,
      "auxiliary_loss_mlp": 0.01024329,
      "balance_loss_clip": 1.00881195,
      "balance_loss_mlp": 1.01787066,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.498659976872526,
      "language_loss": 0.78976071,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.81167406,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 2.7266979217529297
    },
    {
      "auxiliary_loss_clip": 0.01167379,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.00863683,
      "balance_loss_mlp": 1.02067065,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 1.9233049276859338,
      "language_loss": 0.79234463,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81429523,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.592759132385254
    },
    {
      "auxiliary_loss_clip": 0.01168979,
      "auxiliary_loss_mlp": 0.01029975,
      "balance_loss_clip": 1.00907636,
      "balance_loss_mlp": 1.02268577,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 2.144721732998236,
      "language_loss": 0.81613505,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.83812451,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 2.591217041015625
    },
    {
      "auxiliary_loss_clip": 0.0117003,
      "auxiliary_loss_mlp": 0.01023566,
      "balance_loss_clip": 0.89244771,
      "balance_loss_mlp": 1.01660442,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 1.864222995641037,
      "language_loss": 0.74733305,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.76926899,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.733327865600586
    },
    {
      "auxiliary_loss_clip": 0.011634,
      "auxiliary_loss_mlp": 0.01024738,
      "balance_loss_clip": 1.009022,
      "balance_loss_mlp": 1.01781166,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 3.0606260342430875,
      "language_loss": 0.84539866,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.86728001,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 2.6106808185577393
    },
    {
      "auxiliary_loss_clip": 0.01159042,
      "auxiliary_loss_mlp": 0.01027579,
      "balance_loss_clip": 0.97050852,
      "balance_loss_mlp": 1.02064085,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 1.6274342547247693,
      "language_loss": 0.66782463,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.68969083,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 2.6975746154785156
    },
    {
      "auxiliary_loss_clip": 0.01065147,
      "auxiliary_loss_mlp": 0.01001033,
      "balance_loss_clip": 0.97315693,
      "balance_loss_mlp": 0.99944729,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.8166828564400855,
      "language_loss": 0.54527748,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56593931,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.254084348678589
    },
    {
      "auxiliary_loss_clip": 0.01155591,
      "auxiliary_loss_mlp": 0.01027315,
      "balance_loss_clip": 0.96721709,
      "balance_loss_mlp": 1.02053154,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 2.0961403756669954,
      "language_loss": 0.82126492,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.84309399,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.7418880462646484
    },
    {
      "auxiliary_loss_clip": 0.01161613,
      "auxiliary_loss_mlp": 0.01030592,
      "balance_loss_clip": 0.97218108,
      "balance_loss_mlp": 1.02312016,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 2.3728144722178928,
      "language_loss": 0.82940447,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85132647,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 2.5985403060913086
    },
    {
      "auxiliary_loss_clip": 0.01166928,
      "auxiliary_loss_mlp": 0.01028453,
      "balance_loss_clip": 1.01048684,
      "balance_loss_mlp": 1.02149713,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 2.0654940656966128,
      "language_loss": 0.7102496,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73220342,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 2.6435747146606445
    },
    {
      "auxiliary_loss_clip": 0.01177275,
      "auxiliary_loss_mlp": 0.01029136,
      "balance_loss_clip": 1.01333034,
      "balance_loss_mlp": 1.0217452,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 3.8204537790508803,
      "language_loss": 0.7438519,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.76591599,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 2.752187728881836
    },
    {
      "auxiliary_loss_clip": 0.01168676,
      "auxiliary_loss_mlp": 0.01024237,
      "balance_loss_clip": 1.05015683,
      "balance_loss_mlp": 1.01775241,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.049210002577828,
      "language_loss": 0.72437167,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.74630082,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.540818214416504
    },
    {
      "auxiliary_loss_clip": 0.01161726,
      "auxiliary_loss_mlp": 0.01028471,
      "balance_loss_clip": 0.97034001,
      "balance_loss_mlp": 1.02122951,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 1.6313894554231891,
      "language_loss": 0.74159473,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76349664,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.7743563652038574
    },
    {
      "auxiliary_loss_clip": 0.01156839,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.0062021,
      "balance_loss_mlp": 1.01895893,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 1.9639560141782182,
      "language_loss": 0.78717232,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80899572,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.64365816116333
    },
    {
      "auxiliary_loss_clip": 0.01155244,
      "auxiliary_loss_mlp": 0.01030373,
      "balance_loss_clip": 0.93095601,
      "balance_loss_mlp": 1.02353036,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 2.152646381497799,
      "language_loss": 0.71263933,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.73449546,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.6539855003356934
    },
    {
      "auxiliary_loss_clip": 0.01167995,
      "auxiliary_loss_mlp": 0.01029002,
      "balance_loss_clip": 0.97046298,
      "balance_loss_mlp": 1.02131295,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 2.215951883636341,
      "language_loss": 0.80172092,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.82369089,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.6708598136901855
    },
    {
      "auxiliary_loss_clip": 0.01164546,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.01162732,
      "balance_loss_mlp": 1.01631284,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.103132046080869,
      "language_loss": 0.71706784,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.73894238,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 2.6311442852020264
    },
    {
      "auxiliary_loss_clip": 0.01166876,
      "auxiliary_loss_mlp": 0.01025197,
      "balance_loss_clip": 0.93190926,
      "balance_loss_mlp": 1.01802683,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 1.8397041431899528,
      "language_loss": 0.78526366,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.80718446,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.6357717514038086
    },
    {
      "auxiliary_loss_clip": 0.01164802,
      "auxiliary_loss_mlp": 0.01022164,
      "balance_loss_clip": 1.04809129,
      "balance_loss_mlp": 1.015697,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 1.6039651231033567,
      "language_loss": 0.73290682,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75477648,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 2.628063201904297
    },
    {
      "auxiliary_loss_clip": 0.01165412,
      "auxiliary_loss_mlp": 0.0102437,
      "balance_loss_clip": 1.01077211,
      "balance_loss_mlp": 1.01733649,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 4.365343003926625,
      "language_loss": 0.77406073,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.79595852,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 3.544257640838623
    },
    {
      "auxiliary_loss_clip": 0.01165766,
      "auxiliary_loss_mlp": 0.01022731,
      "balance_loss_clip": 1.01194453,
      "balance_loss_mlp": 1.01534641,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 2.058441206214522,
      "language_loss": 0.77750182,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.79938686,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.596633195877075
    },
    {
      "auxiliary_loss_clip": 0.01155833,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 0.92751449,
      "balance_loss_mlp": 1.0201931,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 2.012274596488211,
      "language_loss": 0.72912169,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.75095022,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.903616428375244
    },
    {
      "auxiliary_loss_clip": 0.01165282,
      "auxiliary_loss_mlp": 0.01025713,
      "balance_loss_clip": 1.01066923,
      "balance_loss_mlp": 1.01885831,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 2.037990107263529,
      "language_loss": 0.77430558,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.79621553,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 3.637277126312256
    },
    {
      "auxiliary_loss_clip": 0.0116981,
      "auxiliary_loss_mlp": 0.01026007,
      "balance_loss_clip": 0.93639916,
      "balance_loss_mlp": 1.01869392,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 1.7812641453555786,
      "language_loss": 0.83646786,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.85842597,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 3.569746971130371
    },
    {
      "auxiliary_loss_clip": 0.01171847,
      "auxiliary_loss_mlp": 0.01022719,
      "balance_loss_clip": 1.00991511,
      "balance_loss_mlp": 1.0157243,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 1.8016036358056842,
      "language_loss": 0.73984933,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.76179498,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.7244699001312256
    },
    {
      "auxiliary_loss_clip": 0.01173798,
      "auxiliary_loss_mlp": 0.01027988,
      "balance_loss_clip": 1.01381099,
      "balance_loss_mlp": 1.02020967,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 1.5344936744945334,
      "language_loss": 0.80220944,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82422727,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.6531872749328613
    },
    {
      "auxiliary_loss_clip": 0.01165046,
      "auxiliary_loss_mlp": 0.01026408,
      "balance_loss_clip": 0.93592602,
      "balance_loss_mlp": 1.01920772,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.1654608779794033,
      "language_loss": 0.77404189,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79595649,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 3.6055850982666016
    },
    {
      "auxiliary_loss_clip": 0.01165329,
      "auxiliary_loss_mlp": 0.01027658,
      "balance_loss_clip": 1.01133811,
      "balance_loss_mlp": 1.02043951,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 1.8939989315777086,
      "language_loss": 0.77316296,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79509276,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 2.6666600704193115
    },
    {
      "auxiliary_loss_clip": 0.01169888,
      "auxiliary_loss_mlp": 0.01023395,
      "balance_loss_clip": 1.01202655,
      "balance_loss_mlp": 1.01666868,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 1.4250821195004977,
      "language_loss": 0.74539894,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.76733172,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 2.6938910484313965
    },
    {
      "auxiliary_loss_clip": 0.01174629,
      "auxiliary_loss_mlp": 0.01031083,
      "balance_loss_clip": 0.97485363,
      "balance_loss_mlp": 1.02348387,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 1.7531448787243167,
      "language_loss": 0.74749172,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.76954883,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 2.7391626834869385
    },
    {
      "auxiliary_loss_clip": 0.01168273,
      "auxiliary_loss_mlp": 0.01122192,
      "balance_loss_clip": 0.97182035,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 2.261673620182817,
      "language_loss": 0.66881466,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.69171929,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 2.767219305038452
    },
    {
      "auxiliary_loss_clip": 0.01169732,
      "auxiliary_loss_mlp": 0.01026843,
      "balance_loss_clip": 1.05161309,
      "balance_loss_mlp": 1.02002466,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.435811557751292,
      "language_loss": 0.76912642,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.79109216,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 2.659355401992798
    },
    {
      "auxiliary_loss_clip": 0.01171443,
      "auxiliary_loss_mlp": 0.01024213,
      "balance_loss_clip": 0.97371894,
      "balance_loss_mlp": 1.01663744,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.2428919747503495,
      "language_loss": 0.77396476,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.79592133,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.665325403213501
    },
    {
      "auxiliary_loss_clip": 0.0115881,
      "auxiliary_loss_mlp": 0.01021184,
      "balance_loss_clip": 0.93068099,
      "balance_loss_mlp": 1.01428223,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 2.1342699753447962,
      "language_loss": 0.75581646,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.77761638,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.7532379627227783
    },
    {
      "auxiliary_loss_clip": 0.01164024,
      "auxiliary_loss_mlp": 0.01024389,
      "balance_loss_clip": 0.96943599,
      "balance_loss_mlp": 1.01703966,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.6209411077192009,
      "language_loss": 0.75359279,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77547693,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 2.77107834815979
    },
    {
      "auxiliary_loss_clip": 0.01165777,
      "auxiliary_loss_mlp": 0.0102346,
      "balance_loss_clip": 1.01393533,
      "balance_loss_mlp": 1.01673055,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 2.2372300962555554,
      "language_loss": 0.78208649,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.80397892,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.6960372924804688
    },
    {
      "auxiliary_loss_clip": 0.01165603,
      "auxiliary_loss_mlp": 0.01023221,
      "balance_loss_clip": 0.93531346,
      "balance_loss_mlp": 1.01602709,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 1.8958159106531245,
      "language_loss": 0.6716693,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69355756,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 2.788851737976074
    },
    {
      "auxiliary_loss_clip": 0.01170542,
      "auxiliary_loss_mlp": 0.01122772,
      "balance_loss_clip": 1.04841018,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 2.656840549277873,
      "language_loss": 0.66288555,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.68581873,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.6571731567382812
    },
    {
      "auxiliary_loss_clip": 0.01166585,
      "auxiliary_loss_mlp": 0.0102016,
      "balance_loss_clip": 1.00748587,
      "balance_loss_mlp": 1.01344872,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 1.7020739317762215,
      "language_loss": 0.68833196,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.71019948,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.758697748184204
    },
    {
      "auxiliary_loss_clip": 0.01164964,
      "auxiliary_loss_mlp": 0.0102628,
      "balance_loss_clip": 0.93329686,
      "balance_loss_mlp": 1.01921105,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 2.1315633758046886,
      "language_loss": 0.7248795,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.74679196,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.6614747047424316
    },
    {
      "auxiliary_loss_clip": 0.01162179,
      "auxiliary_loss_mlp": 0.01025292,
      "balance_loss_clip": 0.97245663,
      "balance_loss_mlp": 1.01827669,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.6397013459163916,
      "language_loss": 0.79963148,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.82150626,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.686518669128418
    },
    {
      "auxiliary_loss_clip": 0.01068425,
      "auxiliary_loss_mlp": 0.01001559,
      "balance_loss_clip": 0.9348805,
      "balance_loss_mlp": 0.99993819,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.6998078994001393,
      "language_loss": 0.58213747,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60283732,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.355914831161499
    },
    {
      "auxiliary_loss_clip": 0.01162916,
      "auxiliary_loss_mlp": 0.01026232,
      "balance_loss_clip": 0.97467959,
      "balance_loss_mlp": 1.01880252,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 1.735924010086351,
      "language_loss": 0.81118876,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83308023,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.6832900047302246
    },
    {
      "auxiliary_loss_clip": 0.01168642,
      "auxiliary_loss_mlp": 0.01027931,
      "balance_loss_clip": 0.93333739,
      "balance_loss_mlp": 1.02079618,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 1.7787595573125388,
      "language_loss": 0.68713284,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.70909858,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.8167617321014404
    },
    {
      "auxiliary_loss_clip": 0.01160945,
      "auxiliary_loss_mlp": 0.01029111,
      "balance_loss_clip": 1.01062655,
      "balance_loss_mlp": 1.0221014,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 1.9077950919354163,
      "language_loss": 0.75349271,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.77539325,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.649728298187256
    },
    {
      "auxiliary_loss_clip": 0.01168999,
      "auxiliary_loss_mlp": 0.0103095,
      "balance_loss_clip": 1.04921055,
      "balance_loss_mlp": 1.02422094,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 2.2627093211913385,
      "language_loss": 0.72326589,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.74526536,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 3.4156532287597656
    },
    {
      "auxiliary_loss_clip": 0.0116598,
      "auxiliary_loss_mlp": 0.01024914,
      "balance_loss_clip": 0.9727121,
      "balance_loss_mlp": 1.01833987,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 2.1439730364921523,
      "language_loss": 0.80437267,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82628161,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.690807819366455
    },
    {
      "auxiliary_loss_clip": 0.01167271,
      "auxiliary_loss_mlp": 0.01032318,
      "balance_loss_clip": 1.01230025,
      "balance_loss_mlp": 1.02457845,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 1.7716820866618939,
      "language_loss": 0.70474476,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.72674066,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.7766101360321045
    },
    {
      "auxiliary_loss_clip": 0.01166204,
      "auxiliary_loss_mlp": 0.01022582,
      "balance_loss_clip": 1.01214409,
      "balance_loss_mlp": 1.01567698,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 1.681652992501929,
      "language_loss": 0.74428988,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.76617777,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 3.6730916500091553
    },
    {
      "auxiliary_loss_clip": 0.01174091,
      "auxiliary_loss_mlp": 0.01024767,
      "balance_loss_clip": 1.0116905,
      "balance_loss_mlp": 1.01736426,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 1.911151014610954,
      "language_loss": 0.71189725,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73388588,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 3.6363918781280518
    },
    {
      "auxiliary_loss_clip": 0.01173548,
      "auxiliary_loss_mlp": 0.01032982,
      "balance_loss_clip": 0.8967731,
      "balance_loss_mlp": 1.02551985,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.79934550307467,
      "language_loss": 0.74593663,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76800191,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.7986552715301514
    },
    {
      "auxiliary_loss_clip": 0.01164188,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 0.89997488,
      "balance_loss_mlp": 1.02057445,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.0858815354438476,
      "language_loss": 0.68734765,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.70926017,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 3.5809648036956787
    },
    {
      "auxiliary_loss_clip": 0.01158051,
      "auxiliary_loss_mlp": 0.01024247,
      "balance_loss_clip": 0.89338362,
      "balance_loss_mlp": 1.01716328,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 1.8761681154081,
      "language_loss": 0.66088414,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.68270707,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 2.797194719314575
    },
    {
      "auxiliary_loss_clip": 0.01170935,
      "auxiliary_loss_mlp": 0.01027547,
      "balance_loss_clip": 1.00895762,
      "balance_loss_mlp": 1.01999474,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 1.9063647818133393,
      "language_loss": 0.76565158,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.7876364,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 2.613011360168457
    },
    {
      "auxiliary_loss_clip": 0.01156326,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.00580692,
      "balance_loss_mlp": 1.02237606,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 2.3431633258167603,
      "language_loss": 0.73672539,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.7585814,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 2.5743629932403564
    },
    {
      "auxiliary_loss_clip": 0.01156862,
      "auxiliary_loss_mlp": 0.01028162,
      "balance_loss_clip": 0.93021291,
      "balance_loss_mlp": 1.02080679,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.0011201492355677,
      "language_loss": 0.72472739,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74657768,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.719064712524414
    },
    {
      "auxiliary_loss_clip": 0.01167391,
      "auxiliary_loss_mlp": 0.01021782,
      "balance_loss_clip": 1.01170301,
      "balance_loss_mlp": 1.01513958,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 1.7037316684690165,
      "language_loss": 0.69102114,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.71291292,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.6175365447998047
    },
    {
      "auxiliary_loss_clip": 0.01169825,
      "auxiliary_loss_mlp": 0.01028097,
      "balance_loss_clip": 0.97234923,
      "balance_loss_mlp": 1.02141261,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 2.052493096646598,
      "language_loss": 0.78744745,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.80942667,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.66918683052063
    },
    {
      "auxiliary_loss_clip": 0.01163201,
      "auxiliary_loss_mlp": 0.01024181,
      "balance_loss_clip": 0.97017229,
      "balance_loss_mlp": 1.01697779,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.063934771395369,
      "language_loss": 0.73760635,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.75948018,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 2.672050714492798
    },
    {
      "auxiliary_loss_clip": 0.01160006,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 0.97148144,
      "balance_loss_mlp": 1.01804054,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 2.269827507391957,
      "language_loss": 0.72336066,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74521303,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.709974527359009
    },
    {
      "auxiliary_loss_clip": 0.01173704,
      "auxiliary_loss_mlp": 0.01028419,
      "balance_loss_clip": 0.97046673,
      "balance_loss_mlp": 1.02104926,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 2.110031110953954,
      "language_loss": 0.80869323,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.83071446,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 2.6685116291046143
    },
    {
      "auxiliary_loss_clip": 0.0116388,
      "auxiliary_loss_mlp": 0.01024647,
      "balance_loss_clip": 1.04702294,
      "balance_loss_mlp": 1.01761675,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 1.7548096687039065,
      "language_loss": 0.71703756,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.73892289,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.6427838802337646
    },
    {
      "auxiliary_loss_clip": 0.01158266,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 0.89179063,
      "balance_loss_mlp": 1.01997483,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 2.7330216884299916,
      "language_loss": 0.69601136,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71786219,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.7622573375701904
    },
    {
      "auxiliary_loss_clip": 0.0115675,
      "auxiliary_loss_mlp": 0.01027588,
      "balance_loss_clip": 0.96755803,
      "balance_loss_mlp": 1.02055502,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.8592138581777553,
      "language_loss": 0.84133089,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.86317432,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 2.7119674682617188
    },
    {
      "auxiliary_loss_clip": 0.01154067,
      "auxiliary_loss_mlp": 0.01023555,
      "balance_loss_clip": 0.97084111,
      "balance_loss_mlp": 1.01640201,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 2.1901483320272304,
      "language_loss": 0.66440701,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.68618321,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.745274782180786
    },
    {
      "auxiliary_loss_clip": 0.01166803,
      "auxiliary_loss_mlp": 0.0102834,
      "balance_loss_clip": 0.9728241,
      "balance_loss_mlp": 1.02170873,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 1.6410933117429058,
      "language_loss": 0.83166349,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.85361493,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.7335257530212402
    },
    {
      "auxiliary_loss_clip": 0.01171295,
      "auxiliary_loss_mlp": 0.01023899,
      "balance_loss_clip": 1.01086259,
      "balance_loss_mlp": 1.01623058,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 1.5679589684051565,
      "language_loss": 0.80715811,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.82911003,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.6736526489257812
    },
    {
      "auxiliary_loss_clip": 0.01171885,
      "auxiliary_loss_mlp": 0.01030991,
      "balance_loss_clip": 1.01127148,
      "balance_loss_mlp": 1.02338576,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.7790171210239847,
      "language_loss": 0.76492614,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.78695488,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 2.710193395614624
    },
    {
      "auxiliary_loss_clip": 0.01153298,
      "auxiliary_loss_mlp": 0.01027271,
      "balance_loss_clip": 0.97116148,
      "balance_loss_mlp": 1.02038932,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 1.5017985669923928,
      "language_loss": 0.70227575,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72408146,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.630669593811035
    },
    {
      "auxiliary_loss_clip": 0.01160164,
      "auxiliary_loss_mlp": 0.01022,
      "balance_loss_clip": 0.93338418,
      "balance_loss_mlp": 1.01483846,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 1.9739284581539729,
      "language_loss": 0.7292012,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.75102282,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.7223665714263916
    },
    {
      "auxiliary_loss_clip": 0.01158953,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.00747681,
      "balance_loss_mlp": 1.02105308,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.7595456511000573,
      "language_loss": 0.74883831,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77070856,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.6126291751861572
    },
    {
      "auxiliary_loss_clip": 0.01166425,
      "auxiliary_loss_mlp": 0.01026563,
      "balance_loss_clip": 1.00819468,
      "balance_loss_mlp": 1.01955056,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 1.8837505616793533,
      "language_loss": 0.61320782,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63513768,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 3.5050368309020996
    },
    {
      "auxiliary_loss_clip": 0.01164964,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.00844359,
      "balance_loss_mlp": 1.01903701,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.507873052169503,
      "language_loss": 0.78226382,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80416751,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.720940351486206
    },
    {
      "auxiliary_loss_clip": 0.01155633,
      "auxiliary_loss_mlp": 0.01026696,
      "balance_loss_clip": 0.92892241,
      "balance_loss_mlp": 1.01987123,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.7421960820088846,
      "language_loss": 0.82340342,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.84522676,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.637407064437866
    },
    {
      "auxiliary_loss_clip": 0.01166062,
      "auxiliary_loss_mlp": 0.01021657,
      "balance_loss_clip": 1.00903523,
      "balance_loss_mlp": 1.01490355,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 1.7669472472626073,
      "language_loss": 0.71995085,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74182808,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 3.7265303134918213
    },
    {
      "auxiliary_loss_clip": 0.01161435,
      "auxiliary_loss_mlp": 0.01024298,
      "balance_loss_clip": 0.85620928,
      "balance_loss_mlp": 1.01685596,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 1.9251147197406175,
      "language_loss": 0.63878363,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.66064095,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 3.7138264179229736
    },
    {
      "auxiliary_loss_clip": 0.01059036,
      "auxiliary_loss_mlp": 0.01003495,
      "balance_loss_clip": 0.97163975,
      "balance_loss_mlp": 1.00185001,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7443259258273837,
      "language_loss": 0.58803809,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60866344,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.1010262966156006
    },
    {
      "auxiliary_loss_clip": 0.01171973,
      "auxiliary_loss_mlp": 0.01024542,
      "balance_loss_clip": 1.04884207,
      "balance_loss_mlp": 1.01656127,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 1.7725554090952644,
      "language_loss": 0.88689578,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.90886092,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.621213436126709
    },
    {
      "auxiliary_loss_clip": 0.01170623,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 0.97288406,
      "balance_loss_mlp": 1.01765025,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 2.99437673494826,
      "language_loss": 0.68797493,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70993799,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 3.6006922721862793
    },
    {
      "auxiliary_loss_clip": 0.01168995,
      "auxiliary_loss_mlp": 0.01025023,
      "balance_loss_clip": 1.05008698,
      "balance_loss_mlp": 1.0178678,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.261359572264553,
      "language_loss": 0.75792146,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.77986157,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 2.6368231773376465
    },
    {
      "auxiliary_loss_clip": 0.01169802,
      "auxiliary_loss_mlp": 0.0102838,
      "balance_loss_clip": 0.97243953,
      "balance_loss_mlp": 1.02170694,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 3.0242847056233644,
      "language_loss": 0.72706193,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.7490437,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 2.719367504119873
    },
    {
      "auxiliary_loss_clip": 0.01167288,
      "auxiliary_loss_mlp": 0.01030661,
      "balance_loss_clip": 1.01280951,
      "balance_loss_mlp": 1.02334428,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 2.11030454985203,
      "language_loss": 0.77495003,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.7969296,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.7558059692382812
    },
    {
      "auxiliary_loss_clip": 0.01169212,
      "auxiliary_loss_mlp": 0.01122485,
      "balance_loss_clip": 1.04910398,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.023467870122603,
      "language_loss": 0.71701908,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.73993611,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.6376378536224365
    },
    {
      "auxiliary_loss_clip": 0.01065124,
      "auxiliary_loss_mlp": 0.01001243,
      "balance_loss_clip": 0.97284508,
      "balance_loss_mlp": 0.99960941,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.7782471071193168,
      "language_loss": 0.64318955,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.66385317,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 3.3235673904418945
    },
    {
      "auxiliary_loss_clip": 0.01161062,
      "auxiliary_loss_mlp": 0.01025178,
      "balance_loss_clip": 0.93412066,
      "balance_loss_mlp": 1.01753378,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 1.6660955836263895,
      "language_loss": 0.70466912,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.72653151,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.7402284145355225
    },
    {
      "auxiliary_loss_clip": 0.01162383,
      "auxiliary_loss_mlp": 0.01023888,
      "balance_loss_clip": 1.00810277,
      "balance_loss_mlp": 1.01650023,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 6.097605387190584,
      "language_loss": 0.69833803,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72020078,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 2.6169393062591553
    },
    {
      "auxiliary_loss_clip": 0.01170061,
      "auxiliary_loss_mlp": 0.01026435,
      "balance_loss_clip": 0.93393219,
      "balance_loss_mlp": 1.01925278,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 1.949719205235015,
      "language_loss": 0.74952507,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.77148998,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.7691726684570312
    },
    {
      "auxiliary_loss_clip": 0.01175893,
      "auxiliary_loss_mlp": 0.01027486,
      "balance_loss_clip": 0.97335649,
      "balance_loss_mlp": 1.02008367,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.7983277182610258,
      "language_loss": 0.7126627,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.73469651,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.6686131954193115
    },
    {
      "auxiliary_loss_clip": 0.01171852,
      "auxiliary_loss_mlp": 0.01023889,
      "balance_loss_clip": 0.97262472,
      "balance_loss_mlp": 1.01741862,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.6921938277982265,
      "language_loss": 0.79678082,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.81873828,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 2.628260612487793
    },
    {
      "auxiliary_loss_clip": 0.01158312,
      "auxiliary_loss_mlp": 0.01026356,
      "balance_loss_clip": 0.97146028,
      "balance_loss_mlp": 1.0189476,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 1.7841493311015217,
      "language_loss": 0.80060828,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82245493,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.7094192504882812
    },
    {
      "auxiliary_loss_clip": 0.01062279,
      "auxiliary_loss_mlp": 0.01003648,
      "balance_loss_clip": 0.93716264,
      "balance_loss_mlp": 1.00195491,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.7681381037675021,
      "language_loss": 0.58080113,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60146034,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.2440171241760254
    },
    {
      "auxiliary_loss_clip": 0.01172653,
      "auxiliary_loss_mlp": 0.01028667,
      "balance_loss_clip": 0.9714694,
      "balance_loss_mlp": 1.02093291,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 3.4077068753378867,
      "language_loss": 0.62193471,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.64394796,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.702986240386963
    },
    {
      "auxiliary_loss_clip": 0.01066966,
      "auxiliary_loss_mlp": 0.01116099,
      "balance_loss_clip": 0.93801415,
      "balance_loss_mlp": 0.0,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.8631809625309379,
      "language_loss": 0.52551186,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54734242,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.2151694297790527
    },
    {
      "auxiliary_loss_clip": 0.01163386,
      "auxiliary_loss_mlp": 0.01028554,
      "balance_loss_clip": 1.00987637,
      "balance_loss_mlp": 1.02171433,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 2.171917626060502,
      "language_loss": 0.71614361,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.73806304,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 2.7020928859710693
    },
    {
      "auxiliary_loss_clip": 0.01064841,
      "auxiliary_loss_mlp": 0.0100181,
      "balance_loss_clip": 0.97362602,
      "balance_loss_mlp": 1.00023639,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.6636167266474358,
      "language_loss": 0.61036718,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.63103366,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.210146427154541
    },
    {
      "auxiliary_loss_clip": 0.01165102,
      "auxiliary_loss_mlp": 0.01027773,
      "balance_loss_clip": 1.01138008,
      "balance_loss_mlp": 1.02068341,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 2.343781273293228,
      "language_loss": 0.69362664,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.71555537,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.662916660308838
    },
    {
      "auxiliary_loss_clip": 0.011689,
      "auxiliary_loss_mlp": 0.01022768,
      "balance_loss_clip": 1.04717088,
      "balance_loss_mlp": 1.01547194,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 1.9674634653589176,
      "language_loss": 0.77441847,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.7963351,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 3.3858907222747803
    },
    {
      "auxiliary_loss_clip": 0.0116862,
      "auxiliary_loss_mlp": 0.01030183,
      "balance_loss_clip": 1.01349306,
      "balance_loss_mlp": 1.02275038,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.614260872679887,
      "language_loss": 0.71846527,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.7404533,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.713832378387451
    },
    {
      "auxiliary_loss_clip": 0.01171413,
      "auxiliary_loss_mlp": 0.01021964,
      "balance_loss_clip": 1.01019478,
      "balance_loss_mlp": 1.01461768,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 1.6386910108539334,
      "language_loss": 0.82386976,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.8458035,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.608400583267212
    },
    {
      "auxiliary_loss_clip": 0.01169324,
      "auxiliary_loss_mlp": 0.01023817,
      "balance_loss_clip": 0.93356073,
      "balance_loss_mlp": 1.01693237,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 1.5010394443692616,
      "language_loss": 0.78463978,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80657119,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.697458267211914
    },
    {
      "auxiliary_loss_clip": 0.01156636,
      "auxiliary_loss_mlp": 0.01023635,
      "balance_loss_clip": 0.97141117,
      "balance_loss_mlp": 1.01670897,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 2.924453323757419,
      "language_loss": 0.67358446,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69538718,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.8376846313476562
    },
    {
      "auxiliary_loss_clip": 0.01058715,
      "auxiliary_loss_mlp": 0.0100311,
      "balance_loss_clip": 0.97117114,
      "balance_loss_mlp": 1.00152433,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.8162518125262407,
      "language_loss": 0.67724097,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69785923,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 5.598546028137207
    },
    {
      "auxiliary_loss_clip": 0.01168773,
      "auxiliary_loss_mlp": 0.01033431,
      "balance_loss_clip": 1.04910576,
      "balance_loss_mlp": 1.02599239,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 1.8078327862235222,
      "language_loss": 0.74085724,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.76287919,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.642275094985962
    },
    {
      "auxiliary_loss_clip": 0.01169499,
      "auxiliary_loss_mlp": 0.01027828,
      "balance_loss_clip": 1.00956678,
      "balance_loss_mlp": 1.02029443,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 1.737633706267267,
      "language_loss": 0.71576107,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73773432,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.700253963470459
    },
    {
      "auxiliary_loss_clip": 0.01058452,
      "auxiliary_loss_mlp": 0.0100172,
      "balance_loss_clip": 0.97112864,
      "balance_loss_mlp": 1.00005078,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.7067607961393236,
      "language_loss": 0.5982495,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61885124,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 4.095434665679932
    },
    {
      "auxiliary_loss_clip": 0.01156269,
      "auxiliary_loss_mlp": 0.01026867,
      "balance_loss_clip": 0.93056822,
      "balance_loss_mlp": 1.01940489,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 1.7239748522236884,
      "language_loss": 0.6216507,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.64348209,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 2.783257484436035
    },
    {
      "auxiliary_loss_clip": 0.01166296,
      "auxiliary_loss_mlp": 0.01025666,
      "balance_loss_clip": 0.89519483,
      "balance_loss_mlp": 1.01844239,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 1.8908485271652917,
      "language_loss": 0.82274163,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84466124,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 2.7230944633483887
    },
    {
      "auxiliary_loss_clip": 0.01174501,
      "auxiliary_loss_mlp": 0.01029978,
      "balance_loss_clip": 1.05030191,
      "balance_loss_mlp": 1.02249718,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 1.7635125200075445,
      "language_loss": 0.85022676,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.8722716,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 249.08684515953064
    },
    {
      "auxiliary_loss_clip": 0.01166888,
      "auxiliary_loss_mlp": 0.01025472,
      "balance_loss_clip": 1.01004171,
      "balance_loss_mlp": 1.01860881,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 1.607390689911882,
      "language_loss": 0.75130737,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.77323103,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 2.681439161300659
    },
    {
      "auxiliary_loss_clip": 0.01166729,
      "auxiliary_loss_mlp": 0.0101831,
      "balance_loss_clip": 0.9301936,
      "balance_loss_mlp": 1.01101422,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 1.7286641799811648,
      "language_loss": 0.69258893,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71443927,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 2.8724026679992676
    },
    {
      "auxiliary_loss_clip": 0.01161432,
      "auxiliary_loss_mlp": 0.01023797,
      "balance_loss_clip": 0.93063819,
      "balance_loss_mlp": 1.01663876,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 2.037297213538561,
      "language_loss": 0.79110581,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.81295812,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 2.7092056274414062
    },
    {
      "auxiliary_loss_clip": 0.0115391,
      "auxiliary_loss_mlp": 0.01021466,
      "balance_loss_clip": 0.89094251,
      "balance_loss_mlp": 1.01448917,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 1.7807262294788695,
      "language_loss": 0.74093586,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76268959,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 2.7729671001434326
    },
    {
      "auxiliary_loss_clip": 0.01163819,
      "auxiliary_loss_mlp": 0.01026339,
      "balance_loss_clip": 1.00730407,
      "balance_loss_mlp": 1.01934731,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 1.7092703055983405,
      "language_loss": 0.72696221,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.7488637,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 2.7266480922698975
    },
    {
      "auxiliary_loss_clip": 0.01154653,
      "auxiliary_loss_mlp": 0.01028991,
      "balance_loss_clip": 1.00668776,
      "balance_loss_mlp": 1.02244306,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 2.0548922073301714,
      "language_loss": 0.80590451,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.82774091,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.669433832168579
    },
    {
      "auxiliary_loss_clip": 0.01168587,
      "auxiliary_loss_mlp": 0.01029438,
      "balance_loss_clip": 1.01026571,
      "balance_loss_mlp": 1.0227325,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 1.7123619604687215,
      "language_loss": 0.78012979,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.80211008,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 2.7747952938079834
    },
    {
      "auxiliary_loss_clip": 0.01170372,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 1.00919461,
      "balance_loss_mlp": 1.01879025,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 1.6874933843651985,
      "language_loss": 0.73921382,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.76117611,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.7159814834594727
    },
    {
      "auxiliary_loss_clip": 0.01171738,
      "auxiliary_loss_mlp": 0.01024088,
      "balance_loss_clip": 1.01060748,
      "balance_loss_mlp": 1.01658964,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 1.6039904872683817,
      "language_loss": 0.82961285,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85157108,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.706937551498413
    },
    {
      "auxiliary_loss_clip": 0.01061278,
      "auxiliary_loss_mlp": 0.01002347,
      "balance_loss_clip": 1.00965095,
      "balance_loss_mlp": 1.00066662,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8758292555905414,
      "language_loss": 0.66251791,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68315423,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 3.1428170204162598
    },
    {
      "auxiliary_loss_clip": 0.01168171,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.04839373,
      "balance_loss_mlp": 1.01895177,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 1.7830700516648605,
      "language_loss": 0.76438427,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.78632683,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 2.6442933082580566
    },
    {
      "auxiliary_loss_clip": 0.01155338,
      "auxiliary_loss_mlp": 0.01024782,
      "balance_loss_clip": 0.96838522,
      "balance_loss_mlp": 1.01793599,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 1.6281054479759762,
      "language_loss": 0.70554513,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.7273463,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.7146663665771484
    },
    {
      "auxiliary_loss_clip": 0.01163994,
      "auxiliary_loss_mlp": 0.0102514,
      "balance_loss_clip": 0.93349159,
      "balance_loss_mlp": 1.01823783,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 1.7875655546713796,
      "language_loss": 0.64938217,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.67127347,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.7519052028656006
    },
    {
      "auxiliary_loss_clip": 0.01166801,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 0.97344983,
      "balance_loss_mlp": 1.01608634,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 1.8754701693336862,
      "language_loss": 0.64907664,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.67097998,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.7593581676483154
    },
    {
      "auxiliary_loss_clip": 0.01164874,
      "auxiliary_loss_mlp": 0.01028471,
      "balance_loss_clip": 0.93627775,
      "balance_loss_mlp": 1.02193797,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 2.1385768456286107,
      "language_loss": 0.64825428,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.67018771,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 3.6643426418304443
    },
    {
      "auxiliary_loss_clip": 0.01066686,
      "auxiliary_loss_mlp": 0.01003209,
      "balance_loss_clip": 0.86105883,
      "balance_loss_mlp": 1.00152838,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.6818387961768673,
      "language_loss": 0.51430964,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53500867,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.3500208854675293
    },
    {
      "auxiliary_loss_clip": 0.01158453,
      "auxiliary_loss_mlp": 0.01027284,
      "balance_loss_clip": 0.97044456,
      "balance_loss_mlp": 1.02018499,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 4.088369216329761,
      "language_loss": 0.71264476,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.73450208,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 2.6936709880828857
    },
    {
      "auxiliary_loss_clip": 0.01165116,
      "auxiliary_loss_mlp": 0.01022355,
      "balance_loss_clip": 1.01010275,
      "balance_loss_mlp": 1.01496696,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 2.8452946240699024,
      "language_loss": 0.83455169,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85642636,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.6812679767608643
    },
    {
      "auxiliary_loss_clip": 0.01170098,
      "auxiliary_loss_mlp": 0.01029865,
      "balance_loss_clip": 1.04698324,
      "balance_loss_mlp": 1.02221739,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 3.3228667197168997,
      "language_loss": 0.84434724,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.86634684,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 3.6183021068573
    },
    {
      "auxiliary_loss_clip": 0.01168491,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.00939584,
      "balance_loss_mlp": 1.01929116,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.6784917221101814,
      "language_loss": 0.7217201,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.743662,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.630826473236084
    },
    {
      "auxiliary_loss_clip": 0.01168623,
      "auxiliary_loss_mlp": 0.0102795,
      "balance_loss_clip": 0.97133392,
      "balance_loss_mlp": 1.02071404,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 2.599277773219105,
      "language_loss": 0.65183222,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67379797,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 3.91355562210083
    },
    {
      "auxiliary_loss_clip": 0.01171414,
      "auxiliary_loss_mlp": 0.01032795,
      "balance_loss_clip": 1.01292002,
      "balance_loss_mlp": 1.02580309,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 1.83208940648015,
      "language_loss": 0.71009618,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.73213828,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.728790044784546
    },
    {
      "auxiliary_loss_clip": 0.01165983,
      "auxiliary_loss_mlp": 0.01023588,
      "balance_loss_clip": 1.04763603,
      "balance_loss_mlp": 1.01652157,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 1.8340936224207953,
      "language_loss": 0.7871052,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.80900091,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 3.568702459335327
    },
    {
      "auxiliary_loss_clip": 0.0116796,
      "auxiliary_loss_mlp": 0.01027561,
      "balance_loss_clip": 1.01045561,
      "balance_loss_mlp": 1.02029538,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 2.4280659934907542,
      "language_loss": 0.72776282,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.74971801,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 2.843255043029785
    },
    {
      "auxiliary_loss_clip": 0.01167377,
      "auxiliary_loss_mlp": 0.01028141,
      "balance_loss_clip": 1.01220596,
      "balance_loss_mlp": 1.02062464,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 1.965232543046391,
      "language_loss": 0.73348677,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.7554419,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.6522648334503174
    },
    {
      "auxiliary_loss_clip": 0.01166409,
      "auxiliary_loss_mlp": 0.01023888,
      "balance_loss_clip": 1.01029181,
      "balance_loss_mlp": 1.01648533,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 7.911464510010074,
      "language_loss": 0.74978495,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.77168792,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 2.6385769844055176
    },
    {
      "auxiliary_loss_clip": 0.01167231,
      "auxiliary_loss_mlp": 0.01024377,
      "balance_loss_clip": 0.93584436,
      "balance_loss_mlp": 1.01658106,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 1.8162667127950662,
      "language_loss": 0.77558559,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79750168,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.7017829418182373
    },
    {
      "auxiliary_loss_clip": 0.01163106,
      "auxiliary_loss_mlp": 0.01025559,
      "balance_loss_clip": 0.97098982,
      "balance_loss_mlp": 1.01779544,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 1.850360966515731,
      "language_loss": 0.79428172,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81616831,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 2.676772356033325
    },
    {
      "auxiliary_loss_clip": 0.01170974,
      "auxiliary_loss_mlp": 0.01024347,
      "balance_loss_clip": 1.0131762,
      "balance_loss_mlp": 1.01737344,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 1.9385405696818618,
      "language_loss": 0.6916315,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71358472,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 2.6400535106658936
    },
    {
      "auxiliary_loss_clip": 0.01165078,
      "auxiliary_loss_mlp": 0.01029363,
      "balance_loss_clip": 0.93266773,
      "balance_loss_mlp": 1.02222276,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 1.9573836757641176,
      "language_loss": 0.66993386,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.69187826,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.7388315200805664
    },
    {
      "auxiliary_loss_clip": 0.01167806,
      "auxiliary_loss_mlp": 0.01029108,
      "balance_loss_clip": 1.04712844,
      "balance_loss_mlp": 1.02221477,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 2.0161439130598646,
      "language_loss": 0.85798454,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.87995374,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 2.6451570987701416
    },
    {
      "auxiliary_loss_clip": 0.01168381,
      "auxiliary_loss_mlp": 0.01023198,
      "balance_loss_clip": 0.97254097,
      "balance_loss_mlp": 1.01663852,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.6338643589332176,
      "language_loss": 0.81047243,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83238816,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.6828465461730957
    },
    {
      "auxiliary_loss_clip": 0.01174318,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.01055598,
      "balance_loss_mlp": 1.02305532,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.6349496596993518,
      "language_loss": 0.70463049,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.72667295,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 2.7874815464019775
    },
    {
      "auxiliary_loss_clip": 0.0117322,
      "auxiliary_loss_mlp": 0.01033115,
      "balance_loss_clip": 0.93552631,
      "balance_loss_mlp": 1.02518129,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 2.5301100228110287,
      "language_loss": 0.84962165,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.87168509,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.6692583560943604
    },
    {
      "auxiliary_loss_clip": 0.01158184,
      "auxiliary_loss_mlp": 0.01023641,
      "balance_loss_clip": 0.93425983,
      "balance_loss_mlp": 1.01637566,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 2.097222613331695,
      "language_loss": 0.8783884,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90020663,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 2.7222201824188232
    },
    {
      "auxiliary_loss_clip": 0.0116938,
      "auxiliary_loss_mlp": 0.01026609,
      "balance_loss_clip": 1.05016208,
      "balance_loss_mlp": 1.01951599,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.1261191646204995,
      "language_loss": 0.77541512,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.79737496,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.602065324783325
    },
    {
      "auxiliary_loss_clip": 0.01171332,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 0.97490412,
      "balance_loss_mlp": 1.02084565,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 2.7837108146919514,
      "language_loss": 0.88056421,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.90256584,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.687243700027466
    },
    {
      "auxiliary_loss_clip": 0.01068417,
      "auxiliary_loss_mlp": 0.01003205,
      "balance_loss_clip": 0.89827645,
      "balance_loss_mlp": 1.00160778,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7951856518868844,
      "language_loss": 0.59730244,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61801869,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.3152999877929688
    },
    {
      "auxiliary_loss_clip": 0.01165716,
      "auxiliary_loss_mlp": 0.01122733,
      "balance_loss_clip": 1.01022148,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 1.5093112623964748,
      "language_loss": 0.79282254,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.81570703,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.6598074436187744
    },
    {
      "auxiliary_loss_clip": 0.01158309,
      "auxiliary_loss_mlp": 0.01028508,
      "balance_loss_clip": 0.97032475,
      "balance_loss_mlp": 1.02052093,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 1.716485859912331,
      "language_loss": 0.7275492,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.74941736,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.8324873447418213
    },
    {
      "auxiliary_loss_clip": 0.01060428,
      "auxiliary_loss_mlp": 0.01002415,
      "balance_loss_clip": 1.00902891,
      "balance_loss_mlp": 1.00081778,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7171793244461608,
      "language_loss": 0.59077126,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.61139965,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 4.142330646514893
    },
    {
      "auxiliary_loss_clip": 0.01166166,
      "auxiliary_loss_mlp": 0.01025171,
      "balance_loss_clip": 0.93170428,
      "balance_loss_mlp": 1.01795328,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 2.0038529608957005,
      "language_loss": 0.68153334,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.70344675,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.772702693939209
    },
    {
      "auxiliary_loss_clip": 0.01161166,
      "auxiliary_loss_mlp": 0.0102817,
      "balance_loss_clip": 0.93073702,
      "balance_loss_mlp": 1.02001023,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 2.1853139984770795,
      "language_loss": 0.75025558,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77214897,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.726444959640503
    },
    {
      "auxiliary_loss_clip": 0.01155945,
      "auxiliary_loss_mlp": 0.01022914,
      "balance_loss_clip": 0.85505617,
      "balance_loss_mlp": 1.01554072,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 1.8257292620104357,
      "language_loss": 0.76238036,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.78416896,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 3.8654799461364746
    },
    {
      "auxiliary_loss_clip": 0.01165432,
      "auxiliary_loss_mlp": 0.01024992,
      "balance_loss_clip": 0.93107551,
      "balance_loss_mlp": 1.01695156,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 1.9251594730853974,
      "language_loss": 0.68737966,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.70928395,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.775090217590332
    },
    {
      "auxiliary_loss_clip": 0.01169632,
      "auxiliary_loss_mlp": 0.01024007,
      "balance_loss_clip": 1.0096426,
      "balance_loss_mlp": 1.01645541,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 2.4141683992722895,
      "language_loss": 0.73159993,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.75353628,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 3.668740749359131
    },
    {
      "auxiliary_loss_clip": 0.01158256,
      "auxiliary_loss_mlp": 0.01026572,
      "balance_loss_clip": 0.9323585,
      "balance_loss_mlp": 1.01884687,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 1.8994471660281353,
      "language_loss": 0.80738932,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.82923758,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.7395918369293213
    },
    {
      "auxiliary_loss_clip": 0.01160584,
      "auxiliary_loss_mlp": 0.01021207,
      "balance_loss_clip": 0.89636195,
      "balance_loss_mlp": 1.01468909,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 1.767543248526123,
      "language_loss": 0.66680831,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68862623,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.7560136318206787
    },
    {
      "auxiliary_loss_clip": 0.01171508,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.01190138,
      "balance_loss_mlp": 1.01770425,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 1.8918969180347558,
      "language_loss": 0.74925685,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.77121794,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 3.6259443759918213
    },
    {
      "auxiliary_loss_clip": 0.01170299,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 0.97186816,
      "balance_loss_mlp": 1.02127552,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 2.3914714550114247,
      "language_loss": 0.80649167,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.82848519,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 2.6593005657196045
    },
    {
      "auxiliary_loss_clip": 0.01159059,
      "auxiliary_loss_mlp": 0.01030894,
      "balance_loss_clip": 0.97420585,
      "balance_loss_mlp": 1.02405405,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 1.7795132134507583,
      "language_loss": 0.67309856,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69499803,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.7127256393432617
    },
    {
      "auxiliary_loss_clip": 0.01167981,
      "auxiliary_loss_mlp": 0.01031082,
      "balance_loss_clip": 1.00911653,
      "balance_loss_mlp": 1.0243082,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 1.862553871485776,
      "language_loss": 0.76402879,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.78601944,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.690441370010376
    },
    {
      "auxiliary_loss_clip": 0.01170535,
      "auxiliary_loss_mlp": 0.01122451,
      "balance_loss_clip": 0.93398505,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 1.9755177823276675,
      "language_loss": 0.70780993,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.73073983,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 2.667433261871338
    },
    {
      "auxiliary_loss_clip": 0.01165585,
      "auxiliary_loss_mlp": 0.01027732,
      "balance_loss_clip": 0.9722389,
      "balance_loss_mlp": 1.020612,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 1.7222567331670702,
      "language_loss": 0.73616767,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75810087,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 2.666490077972412
    },
    {
      "auxiliary_loss_clip": 0.01170312,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 0.97224063,
      "balance_loss_mlp": 1.01983309,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 1.9654732870675307,
      "language_loss": 0.95232993,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97430325,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.6989409923553467
    },
    {
      "auxiliary_loss_clip": 0.01163156,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.01116228,
      "balance_loss_mlp": 1.02094018,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.225253272861545,
      "language_loss": 0.71917987,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.74109036,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 2.609726905822754
    },
    {
      "auxiliary_loss_clip": 0.01165579,
      "auxiliary_loss_mlp": 0.01122812,
      "balance_loss_clip": 1.00742054,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 2.6893022720600106,
      "language_loss": 0.76009321,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.7829771,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.8002028465270996
    },
    {
      "auxiliary_loss_clip": 0.01165155,
      "auxiliary_loss_mlp": 0.01023853,
      "balance_loss_clip": 0.96961927,
      "balance_loss_mlp": 1.01706457,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 1.6537910408695495,
      "language_loss": 0.77247906,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.79436916,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 2.7326111793518066
    },
    {
      "auxiliary_loss_clip": 0.01155208,
      "auxiliary_loss_mlp": 0.01122343,
      "balance_loss_clip": 1.00637352,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 1.5526467257290273,
      "language_loss": 0.80976403,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.83253962,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 2.7231218814849854
    },
    {
      "auxiliary_loss_clip": 0.01160868,
      "auxiliary_loss_mlp": 0.01122476,
      "balance_loss_clip": 0.85936522,
      "balance_loss_mlp": 0.0,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 1.6994872682158237,
      "language_loss": 0.72993159,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.75276506,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.855098247528076
    },
    {
      "auxiliary_loss_clip": 0.01168572,
      "auxiliary_loss_mlp": 0.01027868,
      "balance_loss_clip": 1.00709128,
      "balance_loss_mlp": 1.02030969,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 1.6405519705333327,
      "language_loss": 0.71619892,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.73816329,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 2.6666877269744873
    },
    {
      "auxiliary_loss_clip": 0.01163634,
      "auxiliary_loss_mlp": 0.01122259,
      "balance_loss_clip": 0.96991992,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 1.8203127166934556,
      "language_loss": 0.79975998,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.82261896,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.788407802581787
    },
    {
      "auxiliary_loss_clip": 0.01166681,
      "auxiliary_loss_mlp": 0.01032907,
      "balance_loss_clip": 1.01420498,
      "balance_loss_mlp": 1.02596879,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.4678544294700084,
      "language_loss": 0.76331449,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78531039,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.7083935737609863
    },
    {
      "auxiliary_loss_clip": 0.01165166,
      "auxiliary_loss_mlp": 0.01019257,
      "balance_loss_clip": 0.97424251,
      "balance_loss_mlp": 1.01260495,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 1.6940440003831214,
      "language_loss": 0.72632623,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.7481705,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.6562983989715576
    },
    {
      "auxiliary_loss_clip": 0.01169401,
      "auxiliary_loss_mlp": 0.01020717,
      "balance_loss_clip": 1.01100731,
      "balance_loss_mlp": 1.01421666,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 1.6364604962140805,
      "language_loss": 0.78835231,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.8102535,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.6506693363189697
    },
    {
      "auxiliary_loss_clip": 0.01176776,
      "auxiliary_loss_mlp": 0.01028563,
      "balance_loss_clip": 1.01301193,
      "balance_loss_mlp": 1.02133894,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 2.1218305299388134,
      "language_loss": 0.7293188,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.75137216,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.6885998249053955
    },
    {
      "auxiliary_loss_clip": 0.01167571,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 0.97195137,
      "balance_loss_mlp": 1.02348804,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 1.7899963971686093,
      "language_loss": 0.75017858,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77216554,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 3.7103676795959473
    },
    {
      "auxiliary_loss_clip": 0.01168349,
      "auxiliary_loss_mlp": 0.01022286,
      "balance_loss_clip": 0.93177795,
      "balance_loss_mlp": 1.01527095,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 1.7207730733900237,
      "language_loss": 0.77002096,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.79192734,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.812008857727051
    },
    {
      "auxiliary_loss_clip": 0.01170077,
      "auxiliary_loss_mlp": 0.01024671,
      "balance_loss_clip": 1.05218029,
      "balance_loss_mlp": 1.01775122,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 1.8531402938443342,
      "language_loss": 0.76527965,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.78722709,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.841285467147827
    },
    {
      "auxiliary_loss_clip": 0.011774,
      "auxiliary_loss_mlp": 0.01031617,
      "balance_loss_clip": 1.01252639,
      "balance_loss_mlp": 1.02339745,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 2.9973602871560083,
      "language_loss": 0.70630723,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72839737,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 3.8322057723999023
    },
    {
      "auxiliary_loss_clip": 0.01166853,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 0.97000933,
      "balance_loss_mlp": 1.01939559,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.002874380931932,
      "language_loss": 0.73311222,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75504589,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.7949678897857666
    },
    {
      "auxiliary_loss_clip": 0.01170592,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.00978303,
      "balance_loss_mlp": 1.01681185,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 2.428672867341611,
      "language_loss": 0.68617433,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.70812738,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.7767210006713867
    },
    {
      "auxiliary_loss_clip": 0.01166201,
      "auxiliary_loss_mlp": 0.01024654,
      "balance_loss_clip": 1.01153779,
      "balance_loss_mlp": 1.01774883,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 1.7151160876199074,
      "language_loss": 0.71097273,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73288125,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 3.718794822692871
    },
    {
      "auxiliary_loss_clip": 0.01060543,
      "auxiliary_loss_mlp": 0.01115745,
      "balance_loss_clip": 1.00898409,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7430094257910287,
      "language_loss": 0.55604953,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.57781243,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.228944778442383
    },
    {
      "auxiliary_loss_clip": 0.01164117,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 0.93381917,
      "balance_loss_mlp": 1.02487373,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 1.6593387470828946,
      "language_loss": 0.68035394,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.70231342,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 3.676525831222534
    },
    {
      "auxiliary_loss_clip": 0.01166421,
      "auxiliary_loss_mlp": 0.01022439,
      "balance_loss_clip": 1.01000237,
      "balance_loss_mlp": 1.0154233,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 1.850273284991831,
      "language_loss": 0.7293191,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.75120771,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 2.6629679203033447
    },
    {
      "auxiliary_loss_clip": 0.01166711,
      "auxiliary_loss_mlp": 0.0102423,
      "balance_loss_clip": 1.00946808,
      "balance_loss_mlp": 1.01756668,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.699539282736508,
      "language_loss": 0.70608854,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.72799802,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 2.684231996536255
    },
    {
      "auxiliary_loss_clip": 0.01164135,
      "auxiliary_loss_mlp": 0.01027561,
      "balance_loss_clip": 0.93369138,
      "balance_loss_mlp": 1.02069449,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 1.815981511826913,
      "language_loss": 0.74360418,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76552117,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.7457919120788574
    },
    {
      "auxiliary_loss_clip": 0.01060282,
      "auxiliary_loss_mlp": 0.01001745,
      "balance_loss_clip": 1.00891495,
      "balance_loss_mlp": 1.0001595,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.7066402151735466,
      "language_loss": 0.57999814,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.60061836,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 3.3316733837127686
    },
    {
      "auxiliary_loss_clip": 0.01171157,
      "auxiliary_loss_mlp": 0.01028196,
      "balance_loss_clip": 0.97441125,
      "balance_loss_mlp": 1.02169645,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.5738377491889755,
      "language_loss": 0.78842735,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81042087,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 2.741650342941284
    },
    {
      "auxiliary_loss_clip": 0.01167876,
      "auxiliary_loss_mlp": 0.01024704,
      "balance_loss_clip": 1.01277614,
      "balance_loss_mlp": 1.01750422,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 1.7945582637146353,
      "language_loss": 0.74165225,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.76357806,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 2.777409791946411
    },
    {
      "auxiliary_loss_clip": 0.0114218,
      "auxiliary_loss_mlp": 0.01022593,
      "balance_loss_clip": 0.85351163,
      "balance_loss_mlp": 1.01614118,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 1.7303999900962486,
      "language_loss": 0.75998694,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.78163463,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.823646306991577
    },
    {
      "auxiliary_loss_clip": 0.01162491,
      "auxiliary_loss_mlp": 0.01026595,
      "balance_loss_clip": 0.96692044,
      "balance_loss_mlp": 1.01978469,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 1.8122774016739207,
      "language_loss": 0.73216093,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.75405174,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 2.7223169803619385
    },
    {
      "auxiliary_loss_clip": 0.01167091,
      "auxiliary_loss_mlp": 0.01027154,
      "balance_loss_clip": 1.04630888,
      "balance_loss_mlp": 1.01966715,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 1.809856459264776,
      "language_loss": 0.68161559,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70355809,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 2.633488178253174
    },
    {
      "auxiliary_loss_clip": 0.01160257,
      "auxiliary_loss_mlp": 0.01025937,
      "balance_loss_clip": 0.93080831,
      "balance_loss_mlp": 1.01880264,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 1.8550387992686854,
      "language_loss": 0.78207839,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.8039403,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.7522175312042236
    },
    {
      "auxiliary_loss_clip": 0.0115382,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 0.96587348,
      "balance_loss_mlp": 1.01763594,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 1.7296896921069205,
      "language_loss": 0.64262438,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66440701,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.7426490783691406
    },
    {
      "auxiliary_loss_clip": 0.01161261,
      "auxiliary_loss_mlp": 0.01024971,
      "balance_loss_clip": 0.97185349,
      "balance_loss_mlp": 1.01765418,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 1.6785299441216637,
      "language_loss": 0.80848086,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83034319,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 2.6685824394226074
    },
    {
      "auxiliary_loss_clip": 0.01161836,
      "auxiliary_loss_mlp": 0.0102743,
      "balance_loss_clip": 0.93112516,
      "balance_loss_mlp": 1.02062571,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 2.2768324881118476,
      "language_loss": 0.79795885,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.81985152,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.733832597732544
    },
    {
      "auxiliary_loss_clip": 0.01168554,
      "auxiliary_loss_mlp": 0.01025768,
      "balance_loss_clip": 1.00965452,
      "balance_loss_mlp": 1.0178411,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 2.2145337864173493,
      "language_loss": 0.75340796,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77535117,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.586454391479492
    },
    {
      "auxiliary_loss_clip": 0.01158084,
      "auxiliary_loss_mlp": 0.01023514,
      "balance_loss_clip": 0.96898699,
      "balance_loss_mlp": 1.01612937,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 1.769352561746613,
      "language_loss": 0.74516439,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76698041,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.751544952392578
    },
    {
      "auxiliary_loss_clip": 0.01167709,
      "auxiliary_loss_mlp": 0.01025835,
      "balance_loss_clip": 1.0099597,
      "balance_loss_mlp": 1.01934731,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 1.8319557928780401,
      "language_loss": 0.86868721,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.89062262,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.6261563301086426
    },
    {
      "auxiliary_loss_clip": 0.01168212,
      "auxiliary_loss_mlp": 0.01024977,
      "balance_loss_clip": 1.01252019,
      "balance_loss_mlp": 1.01802993,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 1.8617908185283607,
      "language_loss": 0.72397798,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.74590987,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.6344716548919678
    },
    {
      "auxiliary_loss_clip": 0.01155735,
      "auxiliary_loss_mlp": 0.01029483,
      "balance_loss_clip": 0.85388339,
      "balance_loss_mlp": 1.0223155,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.1703213965881476,
      "language_loss": 0.66129321,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.6831454,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 3.7488105297088623
    },
    {
      "auxiliary_loss_clip": 0.011582,
      "auxiliary_loss_mlp": 0.01026181,
      "balance_loss_clip": 0.96784222,
      "balance_loss_mlp": 1.01847434,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 2.1960470198239848,
      "language_loss": 0.86564165,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.88748544,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.6780142784118652
    },
    {
      "auxiliary_loss_clip": 0.01168208,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.01154673,
      "balance_loss_mlp": 1.01761925,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 1.781926872310133,
      "language_loss": 0.72888017,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.75080943,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.702434539794922
    },
    {
      "auxiliary_loss_clip": 0.0115924,
      "auxiliary_loss_mlp": 0.01122591,
      "balance_loss_clip": 0.89270461,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 2.1486226949935063,
      "language_loss": 0.79347479,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.81629312,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 3.8436806201934814
    },
    {
      "auxiliary_loss_clip": 0.01068675,
      "auxiliary_loss_mlp": 0.0100227,
      "balance_loss_clip": 0.9349761,
      "balance_loss_mlp": 1.00058901,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.8103684919255477,
      "language_loss": 0.56282985,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58353931,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 3.3665242195129395
    },
    {
      "auxiliary_loss_clip": 0.01166611,
      "auxiliary_loss_mlp": 0.01025085,
      "balance_loss_clip": 1.00774157,
      "balance_loss_mlp": 1.01820064,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 5.971296251549588,
      "language_loss": 0.75501835,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.77693534,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 3.7537357807159424
    },
    {
      "auxiliary_loss_clip": 0.01066945,
      "auxiliary_loss_mlp": 0.01002392,
      "balance_loss_clip": 0.9358753,
      "balance_loss_mlp": 1.00083029,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7470544702810701,
      "language_loss": 0.64475209,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66544539,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.2175889015197754
    },
    {
      "auxiliary_loss_clip": 0.01167209,
      "auxiliary_loss_mlp": 0.01022337,
      "balance_loss_clip": 1.04811156,
      "balance_loss_mlp": 1.01539278,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 1.9025801443963357,
      "language_loss": 0.8380636,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.85995907,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.672542095184326
    },
    {
      "auxiliary_loss_clip": 0.01166551,
      "auxiliary_loss_mlp": 0.01024582,
      "balance_loss_clip": 1.01254749,
      "balance_loss_mlp": 1.01737833,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.6503448193001446,
      "language_loss": 0.81730461,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.83921593,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 3.776110887527466
    },
    {
      "auxiliary_loss_clip": 0.01170976,
      "auxiliary_loss_mlp": 0.01024652,
      "balance_loss_clip": 0.97328973,
      "balance_loss_mlp": 1.01763082,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 1.5205537426651505,
      "language_loss": 0.69384325,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.71579951,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 2.7277328968048096
    },
    {
      "auxiliary_loss_clip": 0.01168166,
      "auxiliary_loss_mlp": 0.01022457,
      "balance_loss_clip": 1.01278663,
      "balance_loss_mlp": 1.01519978,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.5108519425330618,
      "language_loss": 0.71693587,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.73884207,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.7254841327667236
    },
    {
      "auxiliary_loss_clip": 0.01165585,
      "auxiliary_loss_mlp": 0.0102008,
      "balance_loss_clip": 0.93446851,
      "balance_loss_mlp": 1.01279676,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 1.8685555480322291,
      "language_loss": 0.7720207,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.7938773,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.862213611602783
    },
    {
      "auxiliary_loss_clip": 0.01171463,
      "auxiliary_loss_mlp": 0.01021533,
      "balance_loss_clip": 1.00774646,
      "balance_loss_mlp": 1.01436841,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 1.7324918678432193,
      "language_loss": 0.78442979,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.80635983,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 2.7321131229400635
    },
    {
      "auxiliary_loss_clip": 0.0107897,
      "auxiliary_loss_mlp": 0.01002888,
      "balance_loss_clip": 0.82405841,
      "balance_loss_mlp": 1.00112391,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7073576842956038,
      "language_loss": 0.53850973,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.55932832,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 3.392505168914795
    },
    {
      "auxiliary_loss_clip": 0.01157404,
      "auxiliary_loss_mlp": 0.01025546,
      "balance_loss_clip": 0.89628494,
      "balance_loss_mlp": 1.01878631,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 1.8055977402258256,
      "language_loss": 0.79200077,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81383026,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.75618314743042
    },
    {
      "auxiliary_loss_clip": 0.01171014,
      "auxiliary_loss_mlp": 0.01025225,
      "balance_loss_clip": 0.935583,
      "balance_loss_mlp": 1.01788783,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.6622291730444294,
      "language_loss": 0.70847112,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.73043352,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 2.765577793121338
    },
    {
      "auxiliary_loss_clip": 0.01172692,
      "auxiliary_loss_mlp": 0.01027559,
      "balance_loss_clip": 1.04935193,
      "balance_loss_mlp": 1.01999569,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.2418769109441428,
      "language_loss": 0.76949096,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79149354,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 2.784675359725952
    },
    {
      "auxiliary_loss_clip": 0.01168581,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.04840446,
      "balance_loss_mlp": 1.01884687,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 1.7925038244567306,
      "language_loss": 0.76692253,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.78887051,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 2.65797758102417
    },
    {
      "auxiliary_loss_clip": 0.01168668,
      "auxiliary_loss_mlp": 0.01025331,
      "balance_loss_clip": 1.04889703,
      "balance_loss_mlp": 1.0182023,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 2.9046360496756627,
      "language_loss": 0.77604139,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.79798138,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 2.7295093536376953
    },
    {
      "auxiliary_loss_clip": 0.0116501,
      "auxiliary_loss_mlp": 0.01023096,
      "balance_loss_clip": 0.97010374,
      "balance_loss_mlp": 1.01593411,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 4.86135525009975,
      "language_loss": 0.70392853,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.72580957,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.755000591278076
    },
    {
      "auxiliary_loss_clip": 0.01160193,
      "auxiliary_loss_mlp": 0.01023559,
      "balance_loss_clip": 0.89316833,
      "balance_loss_mlp": 1.01608467,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 2.392095819835577,
      "language_loss": 0.77085483,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.79269236,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 2.773329257965088
    },
    {
      "auxiliary_loss_clip": 0.01164786,
      "auxiliary_loss_mlp": 0.01122051,
      "balance_loss_clip": 1.01233613,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 2.0351958583879672,
      "language_loss": 0.71344054,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.73630893,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.707880973815918
    },
    {
      "auxiliary_loss_clip": 0.01159771,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 0.93202209,
      "balance_loss_mlp": 1.01677346,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 1.9237565298366737,
      "language_loss": 0.70382607,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.7256636,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 2.875016927719116
    },
    {
      "auxiliary_loss_clip": 0.01157412,
      "auxiliary_loss_mlp": 0.01021327,
      "balance_loss_clip": 0.9314872,
      "balance_loss_mlp": 1.0147016,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 1.6968385585342232,
      "language_loss": 0.72017533,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74196267,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.7457525730133057
    },
    {
      "auxiliary_loss_clip": 0.01155272,
      "auxiliary_loss_mlp": 0.01031505,
      "balance_loss_clip": 0.89352071,
      "balance_loss_mlp": 1.02385807,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 1.6930381876213516,
      "language_loss": 0.88208842,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90395617,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.8781838417053223
    },
    {
      "auxiliary_loss_clip": 0.01173466,
      "auxiliary_loss_mlp": 0.01023852,
      "balance_loss_clip": 0.93113971,
      "balance_loss_mlp": 1.01669359,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.7702322312319132,
      "language_loss": 0.78264773,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80462098,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.7453866004943848
    },
    {
      "auxiliary_loss_clip": 0.01171273,
      "auxiliary_loss_mlp": 0.01027065,
      "balance_loss_clip": 1.00989723,
      "balance_loss_mlp": 1.01983833,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 1.730166335682321,
      "language_loss": 0.65162992,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.67361331,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 3.4719507694244385
    },
    {
      "auxiliary_loss_clip": 0.01162368,
      "auxiliary_loss_mlp": 0.01035129,
      "balance_loss_clip": 0.97037327,
      "balance_loss_mlp": 1.02740455,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.81354032218232,
      "language_loss": 0.70475703,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72673202,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.6998963356018066
    },
    {
      "auxiliary_loss_clip": 0.01167746,
      "auxiliary_loss_mlp": 0.01028057,
      "balance_loss_clip": 0.97190118,
      "balance_loss_mlp": 1.02114296,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 1.676117606377304,
      "language_loss": 0.85937911,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.88133717,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.6733970642089844
    },
    {
      "auxiliary_loss_clip": 0.01171513,
      "auxiliary_loss_mlp": 0.01030293,
      "balance_loss_clip": 0.93291533,
      "balance_loss_mlp": 1.02246451,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 3.966985113470793,
      "language_loss": 0.76162934,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.78364742,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 3.741150379180908
    },
    {
      "auxiliary_loss_clip": 0.01165703,
      "auxiliary_loss_mlp": 0.01023075,
      "balance_loss_clip": 0.97217602,
      "balance_loss_mlp": 1.01629448,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 1.8417191434933802,
      "language_loss": 0.7302357,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75212348,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.6061089038848877
    },
    {
      "auxiliary_loss_clip": 0.01157358,
      "auxiliary_loss_mlp": 0.01025217,
      "balance_loss_clip": 0.96922421,
      "balance_loss_mlp": 1.01863694,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.9135106406031808,
      "language_loss": 0.83465534,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.85648108,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.748922109603882
    },
    {
      "auxiliary_loss_clip": 0.01168406,
      "auxiliary_loss_mlp": 0.01029452,
      "balance_loss_clip": 1.01009083,
      "balance_loss_mlp": 1.02153111,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.3732176664538494,
      "language_loss": 0.63275492,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65473348,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 3.724184274673462
    },
    {
      "auxiliary_loss_clip": 0.01164439,
      "auxiliary_loss_mlp": 0.01023476,
      "balance_loss_clip": 0.93192911,
      "balance_loss_mlp": 1.01627314,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 1.8493092959311093,
      "language_loss": 0.74659455,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.76847374,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.7428853511810303
    },
    {
      "auxiliary_loss_clip": 0.01171309,
      "auxiliary_loss_mlp": 0.01024419,
      "balance_loss_clip": 1.01093936,
      "balance_loss_mlp": 1.01720715,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 2.261552826693949,
      "language_loss": 0.71666878,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.73862612,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 3.5940537452697754
    },
    {
      "auxiliary_loss_clip": 0.01169345,
      "auxiliary_loss_mlp": 0.01122497,
      "balance_loss_clip": 1.01282585,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 1.5253810597962931,
      "language_loss": 0.73960471,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.76252317,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.6840500831604004
    },
    {
      "auxiliary_loss_clip": 0.01156526,
      "auxiliary_loss_mlp": 0.01024066,
      "balance_loss_clip": 0.89409983,
      "balance_loss_mlp": 1.01698208,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 1.5662576903572605,
      "language_loss": 0.62229311,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.64409906,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.8451220989227295
    },
    {
      "auxiliary_loss_clip": 0.01168029,
      "auxiliary_loss_mlp": 0.01024207,
      "balance_loss_clip": 1.00959349,
      "balance_loss_mlp": 1.01698911,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 2.0710229511203586,
      "language_loss": 0.73799181,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.75991416,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.6027843952178955
    },
    {
      "auxiliary_loss_clip": 0.01163743,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.0086242,
      "balance_loss_mlp": 1.02126002,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 1.5019576184588146,
      "language_loss": 0.76089507,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.78281665,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 2.7161431312561035
    },
    {
      "auxiliary_loss_clip": 0.01169021,
      "auxiliary_loss_mlp": 0.01022718,
      "balance_loss_clip": 0.97086751,
      "balance_loss_mlp": 1.01502323,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.0840281721853002,
      "language_loss": 0.62180728,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.64372468,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 2.7980823516845703
    },
    {
      "auxiliary_loss_clip": 0.01161792,
      "auxiliary_loss_mlp": 0.01024,
      "balance_loss_clip": 0.93678868,
      "balance_loss_mlp": 1.01705563,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 2.235224925740119,
      "language_loss": 0.89590716,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.91776514,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 2.968857526779175
    },
    {
      "auxiliary_loss_clip": 0.0116985,
      "auxiliary_loss_mlp": 0.01025914,
      "balance_loss_clip": 0.9712801,
      "balance_loss_mlp": 1.01884234,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 2.4121496706501673,
      "language_loss": 0.59750229,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.61945999,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 2.7386744022369385
    },
    {
      "auxiliary_loss_clip": 0.0116485,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 0.96946025,
      "balance_loss_mlp": 1.02137232,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 1.7797539560768543,
      "language_loss": 0.82284397,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84477556,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.694910764694214
    },
    {
      "auxiliary_loss_clip": 0.01164395,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.0066309,
      "balance_loss_mlp": 1.01780725,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 2.003650987807286,
      "language_loss": 0.80269927,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82459611,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 2.705275297164917
    },
    {
      "auxiliary_loss_clip": 0.0117032,
      "auxiliary_loss_mlp": 0.01023452,
      "balance_loss_clip": 1.01040125,
      "balance_loss_mlp": 1.01662111,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.4631155413674697,
      "language_loss": 0.79262835,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81456602,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 2.6936464309692383
    },
    {
      "auxiliary_loss_clip": 0.01158736,
      "auxiliary_loss_mlp": 0.0103056,
      "balance_loss_clip": 1.00619066,
      "balance_loss_mlp": 1.02340722,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.5322099257360045,
      "language_loss": 0.63399065,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.65588361,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 2.810299873352051
    },
    {
      "auxiliary_loss_clip": 0.01169806,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 1.01089001,
      "balance_loss_mlp": 1.02006269,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 1.971587925797797,
      "language_loss": 0.72492158,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.74689436,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.6838529109954834
    },
    {
      "auxiliary_loss_clip": 0.01161416,
      "auxiliary_loss_mlp": 0.01023077,
      "balance_loss_clip": 1.01144099,
      "balance_loss_mlp": 1.01565588,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.4531091555098385,
      "language_loss": 0.67540509,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.69725001,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.674117088317871
    },
    {
      "auxiliary_loss_clip": 0.01164906,
      "auxiliary_loss_mlp": 0.01020103,
      "balance_loss_clip": 1.00974,
      "balance_loss_mlp": 1.01344216,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 3.877060258436539,
      "language_loss": 0.73039901,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.75224912,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.782921552658081
    },
    {
      "auxiliary_loss_clip": 0.01163279,
      "auxiliary_loss_mlp": 0.01123409,
      "balance_loss_clip": 0.89339471,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 17.481572589516535,
      "language_loss": 0.74303573,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.76590264,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.7623250484466553
    },
    {
      "auxiliary_loss_clip": 0.01159283,
      "auxiliary_loss_mlp": 0.01025933,
      "balance_loss_clip": 0.89708382,
      "balance_loss_mlp": 1.01862884,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.717586216529441,
      "language_loss": 0.66559702,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.68744916,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.7629919052124023
    },
    {
      "auxiliary_loss_clip": 0.01164836,
      "auxiliary_loss_mlp": 0.01024511,
      "balance_loss_clip": 0.96727502,
      "balance_loss_mlp": 1.01770711,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 4.209230718229438,
      "language_loss": 0.75205654,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77394998,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.9145302772521973
    },
    {
      "auxiliary_loss_clip": 0.01159116,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 0.89727211,
      "balance_loss_mlp": 1.01799369,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.531987944887198,
      "language_loss": 0.67236632,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.6942029,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 3.7219769954681396
    },
    {
      "auxiliary_loss_clip": 0.01163741,
      "auxiliary_loss_mlp": 0.01027836,
      "balance_loss_clip": 1.04681551,
      "balance_loss_mlp": 1.02081418,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 1.992957085338644,
      "language_loss": 0.72426009,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.74617589,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.7083470821380615
    },
    {
      "auxiliary_loss_clip": 0.01158057,
      "auxiliary_loss_mlp": 0.01025729,
      "balance_loss_clip": 1.00596261,
      "balance_loss_mlp": 1.01882124,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.4663206581177832,
      "language_loss": 0.77671719,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79855508,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.7929482460021973
    },
    {
      "auxiliary_loss_clip": 0.01158635,
      "auxiliary_loss_mlp": 0.01029173,
      "balance_loss_clip": 0.89825296,
      "balance_loss_mlp": 1.02209473,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 1.7860153061977977,
      "language_loss": 0.77613431,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.79801238,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.699820041656494
    },
    {
      "auxiliary_loss_clip": 0.01157554,
      "auxiliary_loss_mlp": 0.01023323,
      "balance_loss_clip": 0.92973858,
      "balance_loss_mlp": 1.01638794,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 2.352740183405966,
      "language_loss": 0.83855319,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.86036193,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 3.703303575515747
    },
    {
      "auxiliary_loss_clip": 0.01169238,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.05014706,
      "balance_loss_mlp": 1.01788592,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 1.9060575451753734,
      "language_loss": 0.76762927,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.78957736,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.6831319332122803
    },
    {
      "auxiliary_loss_clip": 0.01167468,
      "auxiliary_loss_mlp": 0.01027553,
      "balance_loss_clip": 1.01107228,
      "balance_loss_mlp": 1.02054906,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 2.42773924243776,
      "language_loss": 0.65522408,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.67717427,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 3.6631925106048584
    },
    {
      "auxiliary_loss_clip": 0.01069533,
      "auxiliary_loss_mlp": 0.01002591,
      "balance_loss_clip": 0.90021741,
      "balance_loss_mlp": 1.00091004,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7255594889088292,
      "language_loss": 0.60717273,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62789404,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.3618812561035156
    },
    {
      "auxiliary_loss_clip": 0.01162154,
      "auxiliary_loss_mlp": 0.01030728,
      "balance_loss_clip": 1.01003373,
      "balance_loss_mlp": 1.02385581,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 1.7818076790865163,
      "language_loss": 0.81653714,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.83846593,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 3.6107685565948486
    },
    {
      "auxiliary_loss_clip": 0.01153354,
      "auxiliary_loss_mlp": 0.01025804,
      "balance_loss_clip": 0.89210749,
      "balance_loss_mlp": 1.0185802,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 1.990130691489747,
      "language_loss": 0.75125808,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77304965,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.753795623779297
    },
    {
      "auxiliary_loss_clip": 0.01151959,
      "auxiliary_loss_mlp": 0.01022546,
      "balance_loss_clip": 0.92804718,
      "balance_loss_mlp": 1.0150125,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 1.7775608811260197,
      "language_loss": 0.78010571,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.8018508,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.831819534301758
    },
    {
      "auxiliary_loss_clip": 0.01168747,
      "auxiliary_loss_mlp": 0.01030903,
      "balance_loss_clip": 0.93076581,
      "balance_loss_mlp": 1.02375603,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 2.3838694415506603,
      "language_loss": 0.79349232,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.81548887,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.7097620964050293
    },
    {
      "auxiliary_loss_clip": 0.01165861,
      "auxiliary_loss_mlp": 0.01032426,
      "balance_loss_clip": 0.97278827,
      "balance_loss_mlp": 1.02478433,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 1.9071331705237566,
      "language_loss": 0.71735018,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.73933303,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 2.6543478965759277
    },
    {
      "auxiliary_loss_clip": 0.01169304,
      "auxiliary_loss_mlp": 0.01122778,
      "balance_loss_clip": 1.0094049,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 1.7765924496810208,
      "language_loss": 0.73363602,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.75655687,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 2.6643333435058594
    },
    {
      "auxiliary_loss_clip": 0.01170339,
      "auxiliary_loss_mlp": 0.0102555,
      "balance_loss_clip": 1.04832232,
      "balance_loss_mlp": 1.01782477,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.08791787732675,
      "language_loss": 0.59438872,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.61634761,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 2.638617515563965
    },
    {
      "auxiliary_loss_clip": 0.0116313,
      "auxiliary_loss_mlp": 0.01122386,
      "balance_loss_clip": 0.85852295,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 1.865250666233546,
      "language_loss": 0.7601617,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.78301686,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 2.8279426097869873
    },
    {
      "auxiliary_loss_clip": 0.0107979,
      "auxiliary_loss_mlp": 0.01002792,
      "balance_loss_clip": 0.78944921,
      "balance_loss_mlp": 1.00115907,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.8150589838388447,
      "language_loss": 0.59800112,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61882699,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.4000723361968994
    },
    {
      "auxiliary_loss_clip": 0.01162201,
      "auxiliary_loss_mlp": 0.01030519,
      "balance_loss_clip": 0.97064292,
      "balance_loss_mlp": 1.02311003,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 1.6855717169022524,
      "language_loss": 0.71232426,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.7342515,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 3.0863020420074463
    },
    {
      "auxiliary_loss_clip": 0.0115035,
      "auxiliary_loss_mlp": 0.01029336,
      "balance_loss_clip": 0.92856222,
      "balance_loss_mlp": 1.02245116,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.7173790261896347,
      "language_loss": 0.77586037,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.79765725,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 2.7552661895751953
    },
    {
      "auxiliary_loss_clip": 0.01069383,
      "auxiliary_loss_mlp": 0.01001162,
      "balance_loss_clip": 0.89824235,
      "balance_loss_mlp": 0.99955225,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9444845372083067,
      "language_loss": 0.65053797,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.67124343,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 3.3264245986938477
    },
    {
      "auxiliary_loss_clip": 0.0116472,
      "auxiliary_loss_mlp": 0.01023646,
      "balance_loss_clip": 1.04719925,
      "balance_loss_mlp": 1.01707196,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 1.663312014941701,
      "language_loss": 0.80126357,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.82314724,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.9758918285369873
    },
    {
      "auxiliary_loss_clip": 0.01159127,
      "auxiliary_loss_mlp": 0.01027065,
      "balance_loss_clip": 0.96758986,
      "balance_loss_mlp": 1.01956689,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 1.5857603028033804,
      "language_loss": 0.66560233,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.68746424,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.9314990043640137
    },
    {
      "auxiliary_loss_clip": 0.01165306,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.0104413,
      "balance_loss_mlp": 1.01864183,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 1.8746843037438818,
      "language_loss": 0.73487878,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75678587,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 2.948791980743408
    },
    {
      "auxiliary_loss_clip": 0.01147846,
      "auxiliary_loss_mlp": 0.01029187,
      "balance_loss_clip": 0.9659161,
      "balance_loss_mlp": 1.02173901,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.1049193490698297,
      "language_loss": 0.70967484,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.73144519,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.725911855697632
    },
    {
      "auxiliary_loss_clip": 0.01166204,
      "auxiliary_loss_mlp": 0.01024605,
      "balance_loss_clip": 1.04737794,
      "balance_loss_mlp": 1.01805162,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 1.9570349845334474,
      "language_loss": 0.72669661,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.74860477,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.62790584564209
    },
    {
      "auxiliary_loss_clip": 0.01156494,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 0.93096161,
      "balance_loss_mlp": 1.02075851,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 1.7311388355196329,
      "language_loss": 0.74228305,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.76412392,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.759638786315918
    },
    {
      "auxiliary_loss_clip": 0.0115495,
      "auxiliary_loss_mlp": 0.01025254,
      "balance_loss_clip": 0.89550018,
      "balance_loss_mlp": 1.01899838,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 1.930563659296063,
      "language_loss": 0.83308464,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85488665,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 3.7314133644104004
    },
    {
      "auxiliary_loss_clip": 0.01157323,
      "auxiliary_loss_mlp": 0.01027541,
      "balance_loss_clip": 0.93126941,
      "balance_loss_mlp": 1.01956642,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 2.2098225385176136,
      "language_loss": 0.81857717,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.84042579,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.706939220428467
    },
    {
      "auxiliary_loss_clip": 0.01158131,
      "auxiliary_loss_mlp": 0.01028952,
      "balance_loss_clip": 0.97257859,
      "balance_loss_mlp": 1.02144814,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 2.0386896277712974,
      "language_loss": 0.8419978,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86386859,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.7344582080841064
    },
    {
      "auxiliary_loss_clip": 0.01156251,
      "auxiliary_loss_mlp": 0.01025048,
      "balance_loss_clip": 0.85241902,
      "balance_loss_mlp": 1.01786542,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 2.0340784186783183,
      "language_loss": 0.65748966,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.67930269,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.804988384246826
    },
    {
      "auxiliary_loss_clip": 0.01166193,
      "auxiliary_loss_mlp": 0.01024091,
      "balance_loss_clip": 1.00936508,
      "balance_loss_mlp": 1.01661026,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 1.938654302762223,
      "language_loss": 0.74445575,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.76635861,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 3.5802340507507324
    },
    {
      "auxiliary_loss_clip": 0.01169962,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 1.05166817,
      "balance_loss_mlp": 1.01930141,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.7651502052939771,
      "language_loss": 0.87958181,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.90154314,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.6617510318756104
    },
    {
      "auxiliary_loss_clip": 0.01160943,
      "auxiliary_loss_mlp": 0.01027979,
      "balance_loss_clip": 1.00703788,
      "balance_loss_mlp": 1.02075815,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 1.8836193534809273,
      "language_loss": 0.73446119,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.7563504,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 3.7840442657470703
    },
    {
      "auxiliary_loss_clip": 0.01164361,
      "auxiliary_loss_mlp": 0.01032118,
      "balance_loss_clip": 1.01000178,
      "balance_loss_mlp": 1.02337122,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.150578938797219,
      "language_loss": 0.78899318,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.81095803,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.7177841663360596
    },
    {
      "auxiliary_loss_clip": 0.01068391,
      "auxiliary_loss_mlp": 0.01000511,
      "balance_loss_clip": 0.85940629,
      "balance_loss_mlp": 0.99887735,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.8436003592374693,
      "language_loss": 0.57326305,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59395206,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 4.131007671356201
    },
    {
      "auxiliary_loss_clip": 0.01163947,
      "auxiliary_loss_mlp": 0.01022125,
      "balance_loss_clip": 0.97094142,
      "balance_loss_mlp": 1.01549101,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 2.0379872311972824,
      "language_loss": 0.75938946,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.78125018,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.6475400924682617
    },
    {
      "auxiliary_loss_clip": 0.01160257,
      "auxiliary_loss_mlp": 0.01021596,
      "balance_loss_clip": 0.93099058,
      "balance_loss_mlp": 1.01502776,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 2.2077914797594578,
      "language_loss": 0.75682992,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.7786485,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.769158124923706
    },
    {
      "auxiliary_loss_clip": 0.01158056,
      "auxiliary_loss_mlp": 0.010204,
      "balance_loss_clip": 0.93004501,
      "balance_loss_mlp": 1.01409948,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 2.066998667298523,
      "language_loss": 0.85068864,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87247318,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.7663543224334717
    },
    {
      "auxiliary_loss_clip": 0.0117305,
      "auxiliary_loss_mlp": 0.01031932,
      "balance_loss_clip": 0.97302115,
      "balance_loss_mlp": 1.02467275,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 1.6888601629021611,
      "language_loss": 0.81143892,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83348876,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 2.806370973587036
    },
    {
      "auxiliary_loss_clip": 0.01167677,
      "auxiliary_loss_mlp": 0.01027718,
      "balance_loss_clip": 0.97272003,
      "balance_loss_mlp": 1.02060437,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 2.188214402218407,
      "language_loss": 0.77189803,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.79385197,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.665611743927002
    },
    {
      "auxiliary_loss_clip": 0.01167682,
      "auxiliary_loss_mlp": 0.01028339,
      "balance_loss_clip": 0.9723224,
      "balance_loss_mlp": 1.02106392,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 1.9604645474498046,
      "language_loss": 0.74005377,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76201391,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.720409870147705
    },
    {
      "auxiliary_loss_clip": 0.01169989,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.0090487,
      "balance_loss_mlp": 1.01896834,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 2.278990379646486,
      "language_loss": 0.72586548,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74783182,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 2.7489936351776123
    },
    {
      "auxiliary_loss_clip": 0.01160784,
      "auxiliary_loss_mlp": 0.01031479,
      "balance_loss_clip": 0.85593855,
      "balance_loss_mlp": 1.02415371,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 1.850339530461547,
      "language_loss": 0.74743307,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.76935577,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 2.974929094314575
    },
    {
      "auxiliary_loss_clip": 0.01161558,
      "auxiliary_loss_mlp": 0.0102308,
      "balance_loss_clip": 0.96658951,
      "balance_loss_mlp": 1.01606476,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 1.7372933048288812,
      "language_loss": 0.8236506,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84549701,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 2.9894938468933105
    },
    {
      "auxiliary_loss_clip": 0.01165706,
      "auxiliary_loss_mlp": 0.01028234,
      "balance_loss_clip": 1.00952601,
      "balance_loss_mlp": 1.02132273,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.526836097624073,
      "language_loss": 0.76207644,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.78401583,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 2.766641139984131
    },
    {
      "auxiliary_loss_clip": 0.01170188,
      "auxiliary_loss_mlp": 0.01025733,
      "balance_loss_clip": 0.9750452,
      "balance_loss_mlp": 1.01835656,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 1.9822682034200594,
      "language_loss": 0.67435551,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69631469,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.6632843017578125
    },
    {
      "auxiliary_loss_clip": 0.01152584,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 0.93206441,
      "balance_loss_mlp": 1.01826072,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 3.8299305209952523,
      "language_loss": 0.69933176,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.72110921,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 2.7303237915039062
    },
    {
      "auxiliary_loss_clip": 0.01169031,
      "auxiliary_loss_mlp": 0.01030094,
      "balance_loss_clip": 0.97058594,
      "balance_loss_mlp": 1.02259851,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.5294109743080533,
      "language_loss": 0.56595123,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.58794248,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.765042543411255
    },
    {
      "auxiliary_loss_clip": 0.01158583,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 0.96943021,
      "balance_loss_mlp": 1.02057886,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 2.170056796948341,
      "language_loss": 0.75542843,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.77729231,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.6870856285095215
    },
    {
      "auxiliary_loss_clip": 0.01152374,
      "auxiliary_loss_mlp": 0.01031378,
      "balance_loss_clip": 0.88969678,
      "balance_loss_mlp": 1.02391589,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.5350907761708596,
      "language_loss": 0.64106905,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66290659,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 3.2484447956085205
    },
    {
      "auxiliary_loss_clip": 0.01167839,
      "auxiliary_loss_mlp": 0.0102425,
      "balance_loss_clip": 0.97082293,
      "balance_loss_mlp": 1.0169009,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.1213038706929437,
      "language_loss": 0.53068155,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55260241,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.7163872718811035
    },
    {
      "auxiliary_loss_clip": 0.01165123,
      "auxiliary_loss_mlp": 0.01024673,
      "balance_loss_clip": 1.04595757,
      "balance_loss_mlp": 1.01774085,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 2.0447308152236285,
      "language_loss": 0.83779812,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.85969603,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.597080707550049
    },
    {
      "auxiliary_loss_clip": 0.01163976,
      "auxiliary_loss_mlp": 0.0102551,
      "balance_loss_clip": 0.93189716,
      "balance_loss_mlp": 1.01811337,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 2.2950018434325523,
      "language_loss": 0.75753522,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77943015,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 3.6947062015533447
    },
    {
      "auxiliary_loss_clip": 0.011703,
      "auxiliary_loss_mlp": 0.01028244,
      "balance_loss_clip": 0.97495383,
      "balance_loss_mlp": 1.02071619,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.0418279302569813,
      "language_loss": 0.77980906,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80179453,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.678457498550415
    },
    {
      "auxiliary_loss_clip": 0.01168112,
      "auxiliary_loss_mlp": 0.01025602,
      "balance_loss_clip": 1.04721689,
      "balance_loss_mlp": 1.01808274,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.253670850017354,
      "language_loss": 0.79164678,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81358391,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.603030204772949
    },
    {
      "auxiliary_loss_clip": 0.01153564,
      "auxiliary_loss_mlp": 0.01122316,
      "balance_loss_clip": 0.85193717,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 1.9272232366440307,
      "language_loss": 0.65895176,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.6817106,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.8094584941864014
    },
    {
      "auxiliary_loss_clip": 0.01165339,
      "auxiliary_loss_mlp": 0.01024604,
      "balance_loss_clip": 1.01155519,
      "balance_loss_mlp": 1.01723444,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 1.9137665569795694,
      "language_loss": 0.77247,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.7943694,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 3.501753807067871
    },
    {
      "auxiliary_loss_clip": 0.01169542,
      "auxiliary_loss_mlp": 0.01029316,
      "balance_loss_clip": 1.00976813,
      "balance_loss_mlp": 1.02199697,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 1.931855794161472,
      "language_loss": 0.79110539,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.81309402,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.6714742183685303
    },
    {
      "auxiliary_loss_clip": 0.01150444,
      "auxiliary_loss_mlp": 0.01021557,
      "balance_loss_clip": 0.92867273,
      "balance_loss_mlp": 1.01426721,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.6599606433707916,
      "language_loss": 0.82878447,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.85050452,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 2.695391893386841
    },
    {
      "auxiliary_loss_clip": 0.01163423,
      "auxiliary_loss_mlp": 0.01022713,
      "balance_loss_clip": 0.97185254,
      "balance_loss_mlp": 1.01604927,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 1.6010106887633817,
      "language_loss": 0.78328466,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80514604,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 3.7498133182525635
    },
    {
      "auxiliary_loss_clip": 0.01163629,
      "auxiliary_loss_mlp": 0.01028697,
      "balance_loss_clip": 1.00981379,
      "balance_loss_mlp": 1.02136564,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 1.844119457201194,
      "language_loss": 0.69282389,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71474719,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 3.604957103729248
    },
    {
      "auxiliary_loss_clip": 0.01165929,
      "auxiliary_loss_mlp": 0.01023503,
      "balance_loss_clip": 1.00582278,
      "balance_loss_mlp": 1.01595712,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 1.7176847915514364,
      "language_loss": 0.77391583,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.79581016,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.683751344680786
    },
    {
      "auxiliary_loss_clip": 0.01169496,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 0.93298364,
      "balance_loss_mlp": 1.01841807,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 1.8606788708907211,
      "language_loss": 0.80588341,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82783103,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.7299447059631348
    },
    {
      "auxiliary_loss_clip": 0.01166501,
      "auxiliary_loss_mlp": 0.01028379,
      "balance_loss_clip": 1.04599953,
      "balance_loss_mlp": 1.02087462,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 2.5651125860783806,
      "language_loss": 0.74921668,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77116549,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.5234556198120117
    },
    {
      "auxiliary_loss_clip": 0.01163567,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.01029289,
      "balance_loss_mlp": 1.02168787,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 2.677663304926065,
      "language_loss": 0.76294023,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.78486729,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.59053635597229
    },
    {
      "auxiliary_loss_clip": 0.01162278,
      "auxiliary_loss_mlp": 0.01024236,
      "balance_loss_clip": 1.00924015,
      "balance_loss_mlp": 1.01754868,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.3750332683352247,
      "language_loss": 0.82048094,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84234613,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 2.592646360397339
    },
    {
      "auxiliary_loss_clip": 0.01166077,
      "auxiliary_loss_mlp": 0.01123004,
      "balance_loss_clip": 0.92994034,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 2.299050283563028,
      "language_loss": 0.71820122,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.74109209,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.6984972953796387
    },
    {
      "auxiliary_loss_clip": 0.01161083,
      "auxiliary_loss_mlp": 0.0102281,
      "balance_loss_clip": 0.93183672,
      "balance_loss_mlp": 1.01577616,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 2.966013988201879,
      "language_loss": 0.83284134,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.8546803,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 2.6699018478393555
    },
    {
      "auxiliary_loss_clip": 0.01063706,
      "auxiliary_loss_mlp": 0.01002026,
      "balance_loss_clip": 0.9724803,
      "balance_loss_mlp": 1.00046432,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8344527081467745,
      "language_loss": 0.59473085,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61538815,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.282468795776367
    },
    {
      "auxiliary_loss_clip": 0.01165421,
      "auxiliary_loss_mlp": 0.0102965,
      "balance_loss_clip": 0.92962611,
      "balance_loss_mlp": 1.0220449,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 2.2000654726784896,
      "language_loss": 0.69942105,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72137177,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.923171043395996
    },
    {
      "auxiliary_loss_clip": 0.01165396,
      "auxiliary_loss_mlp": 0.01122399,
      "balance_loss_clip": 0.97215801,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.159334367199896,
      "language_loss": 0.81995696,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.84283489,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 2.7636003494262695
    },
    {
      "auxiliary_loss_clip": 0.0115605,
      "auxiliary_loss_mlp": 0.01021216,
      "balance_loss_clip": 1.00740194,
      "balance_loss_mlp": 1.01405156,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 1.869588272424611,
      "language_loss": 0.8567245,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.87849712,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.854491710662842
    },
    {
      "auxiliary_loss_clip": 0.01164405,
      "auxiliary_loss_mlp": 0.01031091,
      "balance_loss_clip": 0.97085923,
      "balance_loss_mlp": 1.02404535,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.039568750416951,
      "language_loss": 0.69620943,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.71816438,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 2.7282135486602783
    },
    {
      "auxiliary_loss_clip": 0.01161102,
      "auxiliary_loss_mlp": 0.01025255,
      "balance_loss_clip": 1.04422474,
      "balance_loss_mlp": 1.01852882,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 1.8586391510066842,
      "language_loss": 0.76402241,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78588593,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.5966646671295166
    },
    {
      "auxiliary_loss_clip": 0.01169546,
      "auxiliary_loss_mlp": 0.01027019,
      "balance_loss_clip": 1.00903606,
      "balance_loss_mlp": 1.01945782,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 1.7124324678229943,
      "language_loss": 0.81599629,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83796203,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.8183908462524414
    },
    {
      "auxiliary_loss_clip": 0.01169432,
      "auxiliary_loss_mlp": 0.01028158,
      "balance_loss_clip": 0.93182206,
      "balance_loss_mlp": 1.02039146,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 1.9478553429131786,
      "language_loss": 0.67988342,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.70185935,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 2.917203903198242
    },
    {
      "auxiliary_loss_clip": 0.0116672,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.0115459,
      "balance_loss_mlp": 1.01900005,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 1.8188598663040971,
      "language_loss": 0.67125309,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.69318104,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 2.6900057792663574
    },
    {
      "auxiliary_loss_clip": 0.01166638,
      "auxiliary_loss_mlp": 0.01026507,
      "balance_loss_clip": 1.00939417,
      "balance_loss_mlp": 1.01965809,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 3.6583442506308455,
      "language_loss": 0.7339462,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.75587767,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.6331300735473633
    },
    {
      "auxiliary_loss_clip": 0.01173285,
      "auxiliary_loss_mlp": 0.01032917,
      "balance_loss_clip": 0.93258196,
      "balance_loss_mlp": 1.025213,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 1.8805694986052899,
      "language_loss": 0.75026327,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.77232528,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 3.564208745956421
    },
    {
      "auxiliary_loss_clip": 0.01159861,
      "auxiliary_loss_mlp": 0.01122265,
      "balance_loss_clip": 0.9321208,
      "balance_loss_mlp": 0.0,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 1.890788567019902,
      "language_loss": 0.85588008,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.87870133,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.689788341522217
    },
    {
      "auxiliary_loss_clip": 0.01167035,
      "auxiliary_loss_mlp": 0.01032498,
      "balance_loss_clip": 1.00728846,
      "balance_loss_mlp": 1.02591181,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.323421581755916,
      "language_loss": 0.80439579,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.8263911,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.6297338008880615
    },
    {
      "auxiliary_loss_clip": 0.01163102,
      "auxiliary_loss_mlp": 0.01021197,
      "balance_loss_clip": 1.00923133,
      "balance_loss_mlp": 1.01383924,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 1.7607516919573756,
      "language_loss": 0.73308563,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.75492859,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.5961201190948486
    },
    {
      "auxiliary_loss_clip": 0.01169477,
      "auxiliary_loss_mlp": 0.01037441,
      "balance_loss_clip": 1.00927067,
      "balance_loss_mlp": 1.0297724,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.6612638146343603,
      "language_loss": 0.72473335,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74680257,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 3.578657627105713
    },
    {
      "auxiliary_loss_clip": 0.01065587,
      "auxiliary_loss_mlp": 0.01003297,
      "balance_loss_clip": 0.93379891,
      "balance_loss_mlp": 1.001652,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.7564190065149581,
      "language_loss": 0.5739212,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.5946101,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.211219549179077
    },
    {
      "auxiliary_loss_clip": 0.01168462,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 1.04903936,
      "balance_loss_mlp": 1.01917148,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.5265368884427875,
      "language_loss": 0.79443443,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81638014,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 2.5769684314727783
    },
    {
      "auxiliary_loss_clip": 0.01167865,
      "auxiliary_loss_mlp": 0.01030115,
      "balance_loss_clip": 0.9347471,
      "balance_loss_mlp": 1.02169847,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 2.352250094465441,
      "language_loss": 0.86019528,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.88217509,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 3.6690845489501953
    },
    {
      "auxiliary_loss_clip": 0.01165082,
      "auxiliary_loss_mlp": 0.01027004,
      "balance_loss_clip": 1.04540634,
      "balance_loss_mlp": 1.02010179,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.6120745901370817,
      "language_loss": 0.73707181,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.75899267,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 3.565905809402466
    },
    {
      "auxiliary_loss_clip": 0.01157522,
      "auxiliary_loss_mlp": 0.01122409,
      "balance_loss_clip": 0.85109508,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 1.8742546574945294,
      "language_loss": 0.81920266,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.84200203,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.8016302585601807
    },
    {
      "auxiliary_loss_clip": 0.01168579,
      "auxiliary_loss_mlp": 0.01029929,
      "balance_loss_clip": 1.01064229,
      "balance_loss_mlp": 1.02305055,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 1.4816389602814033,
      "language_loss": 0.82935715,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.85134226,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.735419750213623
    },
    {
      "auxiliary_loss_clip": 0.01169817,
      "auxiliary_loss_mlp": 0.0112252,
      "balance_loss_clip": 1.04834342,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 4.051524791830979,
      "language_loss": 0.65424454,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.67716789,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.635115146636963
    },
    {
      "auxiliary_loss_clip": 0.01168584,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.00793195,
      "balance_loss_mlp": 1.01837516,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 2.3927225476070415,
      "language_loss": 0.62727904,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.64922726,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.7171828746795654
    },
    {
      "auxiliary_loss_clip": 0.01059708,
      "auxiliary_loss_mlp": 0.01001326,
      "balance_loss_clip": 1.00856471,
      "balance_loss_mlp": 0.99970478,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7138564681768484,
      "language_loss": 0.57423723,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59484762,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.323640823364258
    },
    {
      "auxiliary_loss_clip": 0.01174624,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.05014253,
      "balance_loss_mlp": 1.01880693,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 2.1661800318597613,
      "language_loss": 0.69969863,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72171313,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 2.7436702251434326
    },
    {
      "auxiliary_loss_clip": 0.01155758,
      "auxiliary_loss_mlp": 0.01024439,
      "balance_loss_clip": 0.9697836,
      "balance_loss_mlp": 1.01743877,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 1.690062104468688,
      "language_loss": 0.74884653,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.77064848,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 2.8664708137512207
    },
    {
      "auxiliary_loss_clip": 0.01163826,
      "auxiliary_loss_mlp": 0.01024178,
      "balance_loss_clip": 1.04626143,
      "balance_loss_mlp": 1.01774931,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 1.7091380301942924,
      "language_loss": 0.81729215,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.83917224,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 2.7452547550201416
    },
    {
      "auxiliary_loss_clip": 0.01168381,
      "auxiliary_loss_mlp": 0.01029456,
      "balance_loss_clip": 1.01211643,
      "balance_loss_mlp": 1.02253294,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 2.0237671984464316,
      "language_loss": 0.71065855,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.73263693,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 2.921360492706299
    },
    {
      "auxiliary_loss_clip": 0.01068765,
      "auxiliary_loss_mlp": 0.01000656,
      "balance_loss_clip": 0.82377946,
      "balance_loss_mlp": 0.99871272,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8307968120055068,
      "language_loss": 0.5038051,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.5244993,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.3591644763946533
    },
    {
      "auxiliary_loss_clip": 0.0116061,
      "auxiliary_loss_mlp": 0.0102405,
      "balance_loss_clip": 0.973557,
      "balance_loss_mlp": 1.01761007,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 1.6441828139978416,
      "language_loss": 0.71433848,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73618513,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 2.900179386138916
    },
    {
      "auxiliary_loss_clip": 0.01161319,
      "auxiliary_loss_mlp": 0.01021737,
      "balance_loss_clip": 0.93424892,
      "balance_loss_mlp": 1.01517427,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 2.290657190369938,
      "language_loss": 0.78427541,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80610597,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.751253366470337
    },
    {
      "auxiliary_loss_clip": 0.01165652,
      "auxiliary_loss_mlp": 0.01023726,
      "balance_loss_clip": 0.93472886,
      "balance_loss_mlp": 1.01611459,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 1.9345134998657483,
      "language_loss": 0.64967912,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67157292,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.743396759033203
    },
    {
      "auxiliary_loss_clip": 0.01063001,
      "auxiliary_loss_mlp": 0.01004851,
      "balance_loss_clip": 0.93437642,
      "balance_loss_mlp": 1.00321829,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.8634411097069931,
      "language_loss": 0.64113742,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66181594,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.190113067626953
    },
    {
      "auxiliary_loss_clip": 0.01156493,
      "auxiliary_loss_mlp": 0.0102578,
      "balance_loss_clip": 0.88901991,
      "balance_loss_mlp": 1.01881528,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 2.6801149191313898,
      "language_loss": 0.73908412,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76090688,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.8451859951019287
    },
    {
      "auxiliary_loss_clip": 0.011653,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.01278353,
      "balance_loss_mlp": 1.01834941,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 1.9000527016713544,
      "language_loss": 0.82433581,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.84624171,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.7315709590911865
    },
    {
      "auxiliary_loss_clip": 0.01170644,
      "auxiliary_loss_mlp": 0.01029105,
      "balance_loss_clip": 1.01088786,
      "balance_loss_mlp": 1.02162182,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 1.9614795371559568,
      "language_loss": 0.7963686,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.81836611,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 3.7800893783569336
    },
    {
      "auxiliary_loss_clip": 0.01168555,
      "auxiliary_loss_mlp": 0.01025224,
      "balance_loss_clip": 0.93301207,
      "balance_loss_mlp": 1.01826859,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 2.2926331147990844,
      "language_loss": 0.73821819,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.76015604,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.6916520595550537
    },
    {
      "auxiliary_loss_clip": 0.01163308,
      "auxiliary_loss_mlp": 0.01030162,
      "balance_loss_clip": 0.97065294,
      "balance_loss_mlp": 1.02260447,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 3.1620423269924633,
      "language_loss": 0.7179184,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.73985314,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.641402006149292
    },
    {
      "auxiliary_loss_clip": 0.01155825,
      "auxiliary_loss_mlp": 0.01024513,
      "balance_loss_clip": 0.85283107,
      "balance_loss_mlp": 1.0172565,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 2.3601630398595073,
      "language_loss": 0.79970747,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.82151085,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 3.017786741256714
    },
    {
      "auxiliary_loss_clip": 0.01154938,
      "auxiliary_loss_mlp": 0.01028617,
      "balance_loss_clip": 0.96727002,
      "balance_loss_mlp": 1.02097559,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 2.018128223204553,
      "language_loss": 0.66103673,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.6828723,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.8064491748809814
    },
    {
      "auxiliary_loss_clip": 0.01167657,
      "auxiliary_loss_mlp": 0.01122227,
      "balance_loss_clip": 1.00996065,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 1.9289748562063758,
      "language_loss": 0.91066182,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.93356067,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 3.6332156658172607
    },
    {
      "auxiliary_loss_clip": 0.0106539,
      "auxiliary_loss_mlp": 0.0100118,
      "balance_loss_clip": 0.89688253,
      "balance_loss_mlp": 0.99951154,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.89592946249907,
      "language_loss": 0.63167799,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65234369,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.2150051593780518
    },
    {
      "auxiliary_loss_clip": 0.01162919,
      "auxiliary_loss_mlp": 0.01023445,
      "balance_loss_clip": 1.00838423,
      "balance_loss_mlp": 1.01621807,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 2.0127168620690203,
      "language_loss": 0.80668604,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.82854974,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 2.7278287410736084
    },
    {
      "auxiliary_loss_clip": 0.01164899,
      "auxiliary_loss_mlp": 0.01023144,
      "balance_loss_clip": 0.93313384,
      "balance_loss_mlp": 1.01611042,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 2.4297887039586974,
      "language_loss": 0.66472888,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68660939,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 4.011859655380249
    },
    {
      "auxiliary_loss_clip": 0.01167469,
      "auxiliary_loss_mlp": 0.01030218,
      "balance_loss_clip": 1.010535,
      "balance_loss_mlp": 1.02293396,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.815343041385873,
      "language_loss": 0.70845354,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73043036,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 4.282656669616699
    },
    {
      "auxiliary_loss_clip": 0.01160026,
      "auxiliary_loss_mlp": 0.01028561,
      "balance_loss_clip": 0.93110383,
      "balance_loss_mlp": 1.02126575,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 3.113680117980468,
      "language_loss": 0.71605945,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73794538,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.8883283138275146
    },
    {
      "auxiliary_loss_clip": 0.01156207,
      "auxiliary_loss_mlp": 0.01026288,
      "balance_loss_clip": 0.96971542,
      "balance_loss_mlp": 1.01925206,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 1.8421974924667393,
      "language_loss": 0.71882278,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.74064773,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.881882667541504
    },
    {
      "auxiliary_loss_clip": 0.01158634,
      "auxiliary_loss_mlp": 0.01034874,
      "balance_loss_clip": 0.89579189,
      "balance_loss_mlp": 1.02667189,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.1051200848076608,
      "language_loss": 0.84298265,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86491764,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.705489158630371
    },
    {
      "auxiliary_loss_clip": 0.01168536,
      "auxiliary_loss_mlp": 0.01025431,
      "balance_loss_clip": 1.01111972,
      "balance_loss_mlp": 1.01863921,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 1.6337786573305975,
      "language_loss": 0.77024841,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79218805,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.761610984802246
    },
    {
      "auxiliary_loss_clip": 0.01163162,
      "auxiliary_loss_mlp": 0.01028939,
      "balance_loss_clip": 0.97005355,
      "balance_loss_mlp": 1.02190304,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 2.3516617436131466,
      "language_loss": 0.72598374,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74790472,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.7572128772735596
    },
    {
      "auxiliary_loss_clip": 0.0117064,
      "auxiliary_loss_mlp": 0.01029452,
      "balance_loss_clip": 1.04925632,
      "balance_loss_mlp": 1.02202523,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 3.6604311213425595,
      "language_loss": 0.69029522,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.71229613,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.8547301292419434
    },
    {
      "auxiliary_loss_clip": 0.01165728,
      "auxiliary_loss_mlp": 0.01024534,
      "balance_loss_clip": 1.00944138,
      "balance_loss_mlp": 1.01722956,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 1.842059009611426,
      "language_loss": 0.81476915,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83667183,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 2.732365369796753
    },
    {
      "auxiliary_loss_clip": 0.0116768,
      "auxiliary_loss_mlp": 0.01026474,
      "balance_loss_clip": 0.89382422,
      "balance_loss_mlp": 1.01923776,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 1.998467410341376,
      "language_loss": 0.80068058,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82262212,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 2.858466148376465
    },
    {
      "auxiliary_loss_clip": 0.0116423,
      "auxiliary_loss_mlp": 0.01026754,
      "balance_loss_clip": 1.04687059,
      "balance_loss_mlp": 1.01940179,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 1.6629906318139545,
      "language_loss": 0.79172051,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81363034,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.669809579849243
    },
    {
      "auxiliary_loss_clip": 0.01167753,
      "auxiliary_loss_mlp": 0.01023135,
      "balance_loss_clip": 1.0478555,
      "balance_loss_mlp": 1.01642025,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 1.6719117703444133,
      "language_loss": 0.73240888,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75431776,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 2.772437810897827
    },
    {
      "auxiliary_loss_clip": 0.0116757,
      "auxiliary_loss_mlp": 0.01034215,
      "balance_loss_clip": 1.00818264,
      "balance_loss_mlp": 1.02700889,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 1.7459658437752523,
      "language_loss": 0.72289842,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.74491632,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 2.563359498977661
    },
    {
      "auxiliary_loss_clip": 0.01144234,
      "auxiliary_loss_mlp": 0.01026768,
      "balance_loss_clip": 0.88731462,
      "balance_loss_mlp": 1.02006292,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.2536385854791772,
      "language_loss": 0.5938707,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61558068,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 2.8287510871887207
    },
    {
      "auxiliary_loss_clip": 0.01158677,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 0.96955854,
      "balance_loss_mlp": 1.01800644,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 2.295755338732394,
      "language_loss": 0.85915792,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.88099599,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.6019375324249268
    },
    {
      "auxiliary_loss_clip": 0.01170104,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.01121783,
      "balance_loss_mlp": 1.02051592,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 1.6759136568145026,
      "language_loss": 0.81317896,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.83516163,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.540205717086792
    },
    {
      "auxiliary_loss_clip": 0.0116287,
      "auxiliary_loss_mlp": 0.0102268,
      "balance_loss_clip": 0.96566749,
      "balance_loss_mlp": 1.01541758,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 2.0860215740185004,
      "language_loss": 0.75714451,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.77900004,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.667271375656128
    },
    {
      "auxiliary_loss_clip": 0.01165125,
      "auxiliary_loss_mlp": 0.01024244,
      "balance_loss_clip": 0.97272289,
      "balance_loss_mlp": 1.01711524,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 2.134618527200697,
      "language_loss": 0.65306985,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.67496353,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.6334261894226074
    },
    {
      "auxiliary_loss_clip": 0.0116611,
      "auxiliary_loss_mlp": 0.01020799,
      "balance_loss_clip": 0.97426772,
      "balance_loss_mlp": 1.01441824,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 2.529604946297277,
      "language_loss": 0.83585811,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.85772717,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.708415985107422
    },
    {
      "auxiliary_loss_clip": 0.01152754,
      "auxiliary_loss_mlp": 0.01028413,
      "balance_loss_clip": 0.96763754,
      "balance_loss_mlp": 1.02117062,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 2.5581493927812184,
      "language_loss": 0.73707068,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.75888234,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 3.4744927883148193
    },
    {
      "auxiliary_loss_clip": 0.01164035,
      "auxiliary_loss_mlp": 0.01026201,
      "balance_loss_clip": 0.97107458,
      "balance_loss_mlp": 1.01941824,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.412447388550445,
      "language_loss": 0.71882755,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.74072993,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.6915292739868164
    },
    {
      "auxiliary_loss_clip": 0.01063329,
      "auxiliary_loss_mlp": 0.01004406,
      "balance_loss_clip": 0.97145879,
      "balance_loss_mlp": 1.00272489,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.7984941842039343,
      "language_loss": 0.57908356,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59976089,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.1487820148468018
    },
    {
      "auxiliary_loss_clip": 0.01157617,
      "auxiliary_loss_mlp": 0.01023636,
      "balance_loss_clip": 0.92715782,
      "balance_loss_mlp": 1.01684153,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 2.2068764319830985,
      "language_loss": 0.82978106,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85159361,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.7465109825134277
    },
    {
      "auxiliary_loss_clip": 0.0116867,
      "auxiliary_loss_mlp": 0.01028618,
      "balance_loss_clip": 1.01244962,
      "balance_loss_mlp": 1.02126908,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 1.6928614171249674,
      "language_loss": 0.75766534,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.77963823,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 3.656949758529663
    },
    {
      "auxiliary_loss_clip": 0.01164907,
      "auxiliary_loss_mlp": 0.0103188,
      "balance_loss_clip": 1.00967896,
      "balance_loss_mlp": 1.02453017,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.469678548675013,
      "language_loss": 0.80886412,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.830832,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 2.6701529026031494
    },
    {
      "auxiliary_loss_clip": 0.01153952,
      "auxiliary_loss_mlp": 0.01022397,
      "balance_loss_clip": 0.93376994,
      "balance_loss_mlp": 1.01562858,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 2.6967999293331495,
      "language_loss": 0.732306,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75406945,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 3.6228396892547607
    },
    {
      "auxiliary_loss_clip": 0.01170729,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.00983334,
      "balance_loss_mlp": 1.02083659,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 1.5780173765125471,
      "language_loss": 0.71428758,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73627543,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 2.6173038482666016
    },
    {
      "auxiliary_loss_clip": 0.01172558,
      "auxiliary_loss_mlp": 0.01024748,
      "balance_loss_clip": 0.93368566,
      "balance_loss_mlp": 1.01733887,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 2.597354314010813,
      "language_loss": 0.76259452,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.78456759,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 3.670867443084717
    },
    {
      "auxiliary_loss_clip": 0.01155848,
      "auxiliary_loss_mlp": 0.01022664,
      "balance_loss_clip": 0.96874988,
      "balance_loss_mlp": 1.01550269,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 2.149011298208255,
      "language_loss": 0.77786261,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.79964775,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.7447738647460938
    },
    {
      "auxiliary_loss_clip": 0.01165595,
      "auxiliary_loss_mlp": 0.01027626,
      "balance_loss_clip": 1.01107168,
      "balance_loss_mlp": 1.02089703,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 1.5212697887396274,
      "language_loss": 0.69392943,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.71586168,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.663911819458008
    },
    {
      "auxiliary_loss_clip": 0.0116889,
      "auxiliary_loss_mlp": 0.01122603,
      "balance_loss_clip": 1.01101494,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 1.9074027510129383,
      "language_loss": 0.76372874,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.78664362,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.6761653423309326
    },
    {
      "auxiliary_loss_clip": 0.01162233,
      "auxiliary_loss_mlp": 0.01030129,
      "balance_loss_clip": 0.81907409,
      "balance_loss_mlp": 1.02248156,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 1.824229645663038,
      "language_loss": 0.75072861,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.77265215,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.809164047241211
    },
    {
      "auxiliary_loss_clip": 0.01164117,
      "auxiliary_loss_mlp": 0.01025888,
      "balance_loss_clip": 0.97162855,
      "balance_loss_mlp": 1.01981401,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 1.7903888867384188,
      "language_loss": 0.64912325,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.67102325,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.764599084854126
    },
    {
      "auxiliary_loss_clip": 0.01171871,
      "auxiliary_loss_mlp": 0.01022069,
      "balance_loss_clip": 1.00905085,
      "balance_loss_mlp": 1.0152086,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 2.720789186202166,
      "language_loss": 0.68975317,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71169257,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.6230618953704834
    },
    {
      "auxiliary_loss_clip": 0.01163252,
      "auxiliary_loss_mlp": 0.01121963,
      "balance_loss_clip": 0.96942902,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 1.443731166865201,
      "language_loss": 0.78580844,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.80866057,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 2.6400749683380127
    },
    {
      "auxiliary_loss_clip": 0.01163376,
      "auxiliary_loss_mlp": 0.01022407,
      "balance_loss_clip": 1.04605663,
      "balance_loss_mlp": 1.01564777,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 1.8198447002877087,
      "language_loss": 0.76866043,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.79051828,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 2.5311644077301025
    },
    {
      "auxiliary_loss_clip": 0.01155016,
      "auxiliary_loss_mlp": 0.0102411,
      "balance_loss_clip": 0.93143511,
      "balance_loss_mlp": 1.01720476,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 1.8510724039393225,
      "language_loss": 0.78178191,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.80357325,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 2.683535575866699
    },
    {
      "auxiliary_loss_clip": 0.01169349,
      "auxiliary_loss_mlp": 0.01026126,
      "balance_loss_clip": 1.04894805,
      "balance_loss_mlp": 1.01925933,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 1.6956904050738533,
      "language_loss": 0.83696765,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85892242,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 2.554237127304077
    },
    {
      "auxiliary_loss_clip": 0.01171177,
      "auxiliary_loss_mlp": 0.01024586,
      "balance_loss_clip": 0.93163717,
      "balance_loss_mlp": 1.01686716,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.256600351794879,
      "language_loss": 0.78942919,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.81138682,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.723588466644287
    },
    {
      "auxiliary_loss_clip": 0.01165629,
      "auxiliary_loss_mlp": 0.01022675,
      "balance_loss_clip": 1.00776172,
      "balance_loss_mlp": 1.01547432,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 2.0307228617151667,
      "language_loss": 0.6762594,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.69814241,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 2.6733014583587646
    },
    {
      "auxiliary_loss_clip": 0.01167142,
      "auxiliary_loss_mlp": 0.01026696,
      "balance_loss_clip": 0.97067922,
      "balance_loss_mlp": 1.01920378,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 1.904171733065994,
      "language_loss": 0.77933848,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.8012768,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 2.6494553089141846
    },
    {
      "auxiliary_loss_clip": 0.01159279,
      "auxiliary_loss_mlp": 0.01024703,
      "balance_loss_clip": 0.93452501,
      "balance_loss_mlp": 1.01796532,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 1.9500347799172228,
      "language_loss": 0.74734104,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.76918089,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.6982057094573975
    },
    {
      "auxiliary_loss_clip": 0.01167011,
      "auxiliary_loss_mlp": 0.01028792,
      "balance_loss_clip": 1.00803828,
      "balance_loss_mlp": 1.0222919,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 1.822129180547346,
      "language_loss": 0.6889599,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71091795,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 2.7196030616760254
    },
    {
      "auxiliary_loss_clip": 0.01161039,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 0.93081903,
      "balance_loss_mlp": 1.01779258,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 1.9628736627310865,
      "language_loss": 0.79850978,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.82036376,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.636723756790161
    },
    {
      "auxiliary_loss_clip": 0.01167278,
      "auxiliary_loss_mlp": 0.0102424,
      "balance_loss_clip": 0.97154111,
      "balance_loss_mlp": 1.01707268,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 1.677796071283692,
      "language_loss": 0.79628628,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.81820142,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 3.78293514251709
    },
    {
      "auxiliary_loss_clip": 0.01163601,
      "auxiliary_loss_mlp": 0.01024735,
      "balance_loss_clip": 0.93204045,
      "balance_loss_mlp": 1.01768708,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.5079746972860595,
      "language_loss": 0.84466863,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.86655194,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.737521171569824
    },
    {
      "auxiliary_loss_clip": 0.01156606,
      "auxiliary_loss_mlp": 0.0102629,
      "balance_loss_clip": 1.006019,
      "balance_loss_mlp": 1.01928997,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.620490996111539,
      "language_loss": 0.6675036,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.68933254,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.585909128189087
    },
    {
      "auxiliary_loss_clip": 0.01167921,
      "auxiliary_loss_mlp": 0.01024697,
      "balance_loss_clip": 0.97548914,
      "balance_loss_mlp": 1.01755071,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.4719601851392554,
      "language_loss": 0.80256176,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.82448792,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.6330649852752686
    },
    {
      "auxiliary_loss_clip": 0.01159615,
      "auxiliary_loss_mlp": 0.01023984,
      "balance_loss_clip": 0.8956601,
      "balance_loss_mlp": 1.01677787,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 2.7349335503341483,
      "language_loss": 0.69813037,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.71996641,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.623090982437134
    },
    {
      "auxiliary_loss_clip": 0.01167329,
      "auxiliary_loss_mlp": 0.01024526,
      "balance_loss_clip": 1.046749,
      "balance_loss_mlp": 1.01783299,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.701793342282133,
      "language_loss": 0.73784447,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.759763,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 3.549546241760254
    },
    {
      "auxiliary_loss_clip": 0.01169582,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.01086533,
      "balance_loss_mlp": 1.01744902,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 1.805778338478746,
      "language_loss": 0.80607051,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.828013,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 2.598569393157959
    },
    {
      "auxiliary_loss_clip": 0.01158728,
      "auxiliary_loss_mlp": 0.01025424,
      "balance_loss_clip": 1.00738335,
      "balance_loss_mlp": 1.01810741,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 2.10357385703202,
      "language_loss": 0.73610061,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.75794214,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.605057716369629
    },
    {
      "auxiliary_loss_clip": 0.01166413,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 1.00922835,
      "balance_loss_mlp": 1.01860619,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.6148255783043541,
      "language_loss": 0.75200206,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77392113,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 3.5301151275634766
    },
    {
      "auxiliary_loss_clip": 0.01163618,
      "auxiliary_loss_mlp": 0.01025847,
      "balance_loss_clip": 0.93293703,
      "balance_loss_mlp": 1.01834321,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 1.6929602382229778,
      "language_loss": 0.72466636,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74656105,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 3.7218334674835205
    },
    {
      "auxiliary_loss_clip": 0.0116583,
      "auxiliary_loss_mlp": 0.01029169,
      "balance_loss_clip": 1.01188874,
      "balance_loss_mlp": 1.02201009,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 1.8113240289463812,
      "language_loss": 0.70425975,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.7262097,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.620643377304077
    },
    {
      "auxiliary_loss_clip": 0.01158275,
      "auxiliary_loss_mlp": 0.01029284,
      "balance_loss_clip": 0.96849626,
      "balance_loss_mlp": 1.02158904,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.6059876057628926,
      "language_loss": 0.69354528,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71542084,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.6283771991729736
    },
    {
      "auxiliary_loss_clip": 0.01167034,
      "auxiliary_loss_mlp": 0.01023068,
      "balance_loss_clip": 0.97317326,
      "balance_loss_mlp": 1.01599884,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 1.7390399648275723,
      "language_loss": 0.88263422,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.90453523,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.5908517837524414
    },
    {
      "auxiliary_loss_clip": 0.01163144,
      "auxiliary_loss_mlp": 0.01022059,
      "balance_loss_clip": 1.04830599,
      "balance_loss_mlp": 1.0153389,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 1.6627710182951256,
      "language_loss": 0.81509781,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.83694988,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.550320863723755
    },
    {
      "auxiliary_loss_clip": 0.01159588,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 0.93636143,
      "balance_loss_mlp": 1.02056527,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 2.2818211069583874,
      "language_loss": 0.75449622,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.77636826,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.616266965866089
    },
    {
      "auxiliary_loss_clip": 0.01170835,
      "auxiliary_loss_mlp": 0.01029095,
      "balance_loss_clip": 0.9763813,
      "balance_loss_mlp": 1.02232718,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 2.06048581199858,
      "language_loss": 0.81903303,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.84103233,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.567434072494507
    },
    {
      "auxiliary_loss_clip": 0.01167662,
      "auxiliary_loss_mlp": 0.01026761,
      "balance_loss_clip": 1.00967038,
      "balance_loss_mlp": 1.01939344,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 1.8042720057726898,
      "language_loss": 0.67615116,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.69809532,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 2.573303461074829
    },
    {
      "auxiliary_loss_clip": 0.0115578,
      "auxiliary_loss_mlp": 0.01029033,
      "balance_loss_clip": 0.88977909,
      "balance_loss_mlp": 1.02220225,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 1.7249301203225935,
      "language_loss": 0.6530031,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67485118,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 2.7543115615844727
    },
    {
      "auxiliary_loss_clip": 0.01159057,
      "auxiliary_loss_mlp": 0.01122722,
      "balance_loss_clip": 0.96792483,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 1.7272221245876331,
      "language_loss": 0.83919591,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.8620137,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 2.6734609603881836
    },
    {
      "auxiliary_loss_clip": 0.0116517,
      "auxiliary_loss_mlp": 0.01029433,
      "balance_loss_clip": 1.0087285,
      "balance_loss_mlp": 1.02244985,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 1.7665492157204767,
      "language_loss": 0.79721516,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81916118,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 2.6708199977874756
    },
    {
      "auxiliary_loss_clip": 0.01169039,
      "auxiliary_loss_mlp": 0.01032142,
      "balance_loss_clip": 1.0468421,
      "balance_loss_mlp": 1.02474451,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 2.516626763684293,
      "language_loss": 0.72829962,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75031137,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.5968990325927734
    },
    {
      "auxiliary_loss_clip": 0.01169856,
      "auxiliary_loss_mlp": 0.01024004,
      "balance_loss_clip": 0.97463918,
      "balance_loss_mlp": 1.01736379,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 1.8330011687480292,
      "language_loss": 0.87770343,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.89964199,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.620774269104004
    },
    {
      "auxiliary_loss_clip": 0.01165546,
      "auxiliary_loss_mlp": 0.01027971,
      "balance_loss_clip": 0.89262259,
      "balance_loss_mlp": 1.01983523,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 2.8942675105998625,
      "language_loss": 0.62834644,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.65028167,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 2.7794625759124756
    },
    {
      "auxiliary_loss_clip": 0.01159159,
      "auxiliary_loss_mlp": 0.01123102,
      "balance_loss_clip": 0.93261856,
      "balance_loss_mlp": 0.0,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 1.8411677817469903,
      "language_loss": 0.75852156,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.78134412,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.658600091934204
    },
    {
      "auxiliary_loss_clip": 0.01165543,
      "auxiliary_loss_mlp": 0.01022272,
      "balance_loss_clip": 0.96872097,
      "balance_loss_mlp": 1.01573968,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 1.7017188247277764,
      "language_loss": 0.84143662,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.86331481,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.6157543659210205
    },
    {
      "auxiliary_loss_clip": 0.0116782,
      "auxiliary_loss_mlp": 0.01024371,
      "balance_loss_clip": 0.89582253,
      "balance_loss_mlp": 1.01766539,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 2.546471677954222,
      "language_loss": 0.8220582,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84398007,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.727090835571289
    },
    {
      "auxiliary_loss_clip": 0.01168798,
      "auxiliary_loss_mlp": 0.01025804,
      "balance_loss_clip": 1.05026042,
      "balance_loss_mlp": 1.01854753,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 2.0105573206588327,
      "language_loss": 0.70545357,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.72739959,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.511021614074707
    },
    {
      "auxiliary_loss_clip": 0.01163977,
      "auxiliary_loss_mlp": 0.01025132,
      "balance_loss_clip": 0.97205764,
      "balance_loss_mlp": 1.01819086,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 3.4748428245931318,
      "language_loss": 0.86451048,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.88640153,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 3.3788342475891113
    },
    {
      "auxiliary_loss_clip": 0.01171594,
      "auxiliary_loss_mlp": 0.01122691,
      "balance_loss_clip": 1.00935984,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 1.827864628965577,
      "language_loss": 0.65538275,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.67832553,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.6038243770599365
    },
    {
      "auxiliary_loss_clip": 0.01165333,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 0.93357968,
      "balance_loss_mlp": 1.01955867,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 1.8623917528180773,
      "language_loss": 0.87222219,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.89414167,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.7104740142822266
    },
    {
      "auxiliary_loss_clip": 0.01168709,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 0.97452134,
      "balance_loss_mlp": 1.0189904,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 1.6657010627505544,
      "language_loss": 0.80336905,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.82531363,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.651154041290283
    },
    {
      "auxiliary_loss_clip": 0.01065118,
      "auxiliary_loss_mlp": 0.010032,
      "balance_loss_clip": 0.937105,
      "balance_loss_mlp": 1.00154328,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.8685381318027584,
      "language_loss": 0.5646019,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58528507,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 4.251953125
    },
    {
      "auxiliary_loss_clip": 0.01168759,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 0.97304606,
      "balance_loss_mlp": 1.01918435,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 1.8409194001332205,
      "language_loss": 0.77635747,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.79830647,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.6032326221466064
    },
    {
      "auxiliary_loss_clip": 0.01163045,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 0.97165692,
      "balance_loss_mlp": 1.02210474,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 1.6541485121037218,
      "language_loss": 0.67519081,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69711721,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.568554162979126
    },
    {
      "auxiliary_loss_clip": 0.01163869,
      "auxiliary_loss_mlp": 0.01023569,
      "balance_loss_clip": 0.93039715,
      "balance_loss_mlp": 1.01696491,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 1.700489982351643,
      "language_loss": 0.73568821,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75756258,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 3.655773639678955
    },
    {
      "auxiliary_loss_clip": 0.01159881,
      "auxiliary_loss_mlp": 0.01026216,
      "balance_loss_clip": 0.93399787,
      "balance_loss_mlp": 1.01899445,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 2.161056254900189,
      "language_loss": 0.84871399,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.87057495,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 3.6442716121673584
    },
    {
      "auxiliary_loss_clip": 0.01169891,
      "auxiliary_loss_mlp": 0.01025993,
      "balance_loss_clip": 1.01169157,
      "balance_loss_mlp": 1.01914191,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 2.2149421917812733,
      "language_loss": 0.73314047,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.7550993,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.567681312561035
    },
    {
      "auxiliary_loss_clip": 0.01163916,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 0.97317362,
      "balance_loss_mlp": 1.02249575,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 2.2205958816317817,
      "language_loss": 0.73080826,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75275105,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.5772855281829834
    },
    {
      "auxiliary_loss_clip": 0.01148286,
      "auxiliary_loss_mlp": 0.01024457,
      "balance_loss_clip": 0.88912773,
      "balance_loss_mlp": 1.01778746,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 1.9328255108833317,
      "language_loss": 0.68784076,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.7095682,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.686781644821167
    },
    {
      "auxiliary_loss_clip": 0.01166709,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 0.89829004,
      "balance_loss_mlp": 1.02233231,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 1.7261892127069236,
      "language_loss": 0.77916646,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.80113208,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.6893017292022705
    },
    {
      "auxiliary_loss_clip": 0.01157079,
      "auxiliary_loss_mlp": 0.01024596,
      "balance_loss_clip": 0.88753939,
      "balance_loss_mlp": 1.01774168,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 3.2627133886070068,
      "language_loss": 0.69372702,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71554381,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 2.7173776626586914
    },
    {
      "auxiliary_loss_clip": 0.01059441,
      "auxiliary_loss_mlp": 0.01000691,
      "balance_loss_clip": 1.00862765,
      "balance_loss_mlp": 0.99912971,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8785261094226915,
      "language_loss": 0.66361654,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.68421787,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.2009646892547607
    },
    {
      "auxiliary_loss_clip": 0.01164028,
      "auxiliary_loss_mlp": 0.01122161,
      "balance_loss_clip": 1.01005232,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 1.6418613101049777,
      "language_loss": 0.65065134,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.67351317,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 2.573272705078125
    },
    {
      "auxiliary_loss_clip": 0.01171015,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 0.9715904,
      "balance_loss_mlp": 1.02214086,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 3.4787010747688307,
      "language_loss": 0.86645114,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.88846159,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 2.7169039249420166
    },
    {
      "auxiliary_loss_clip": 0.01059546,
      "auxiliary_loss_mlp": 0.01001726,
      "balance_loss_clip": 1.00847697,
      "balance_loss_mlp": 1.00012827,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.8359417200835729,
      "language_loss": 0.60344094,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62405366,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.150742292404175
    },
    {
      "auxiliary_loss_clip": 0.0116749,
      "auxiliary_loss_mlp": 0.01025992,
      "balance_loss_clip": 0.93057173,
      "balance_loss_mlp": 1.01905441,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 1.734104140050404,
      "language_loss": 0.7641421,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.7860769,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 2.653679847717285
    },
    {
      "auxiliary_loss_clip": 0.01161282,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 0.9704982,
      "balance_loss_mlp": 1.01747525,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 1.9546301643739052,
      "language_loss": 0.70787072,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.72973031,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 2.5900583267211914
    },
    {
      "auxiliary_loss_clip": 0.01156761,
      "auxiliary_loss_mlp": 0.0102651,
      "balance_loss_clip": 0.93058079,
      "balance_loss_mlp": 1.01959932,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 2.020532084440526,
      "language_loss": 0.83960497,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86143762,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.723597526550293
    },
    {
      "auxiliary_loss_clip": 0.01161314,
      "auxiliary_loss_mlp": 0.01026617,
      "balance_loss_clip": 0.9313904,
      "balance_loss_mlp": 1.01899922,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 1.8330524768010967,
      "language_loss": 0.73983604,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.76171535,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.7233591079711914
    },
    {
      "auxiliary_loss_clip": 0.01163561,
      "auxiliary_loss_mlp": 0.01022089,
      "balance_loss_clip": 0.97320694,
      "balance_loss_mlp": 1.01481104,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.5391212155006555,
      "language_loss": 0.77736735,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.79922384,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 2.6771841049194336
    },
    {
      "auxiliary_loss_clip": 0.01171353,
      "auxiliary_loss_mlp": 0.01025832,
      "balance_loss_clip": 1.04934251,
      "balance_loss_mlp": 1.01792824,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 1.603532951873198,
      "language_loss": 0.76557612,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.78754795,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.685519218444824
    },
    {
      "auxiliary_loss_clip": 0.01169326,
      "auxiliary_loss_mlp": 0.01020431,
      "balance_loss_clip": 0.89243662,
      "balance_loss_mlp": 1.01337957,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 3.445812409125151,
      "language_loss": 0.71954131,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.74143887,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.8170108795166016
    },
    {
      "auxiliary_loss_clip": 0.011624,
      "auxiliary_loss_mlp": 0.01024733,
      "balance_loss_clip": 1.00891435,
      "balance_loss_mlp": 1.0175066,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.5439020443591682,
      "language_loss": 0.76335537,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.7852267,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.598479747772217
    },
    {
      "auxiliary_loss_clip": 0.01157421,
      "auxiliary_loss_mlp": 0.01031288,
      "balance_loss_clip": 0.89312583,
      "balance_loss_mlp": 1.02333045,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 1.917126923547288,
      "language_loss": 0.64350426,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66539139,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 3.543980598449707
    },
    {
      "auxiliary_loss_clip": 0.01167118,
      "auxiliary_loss_mlp": 0.01023744,
      "balance_loss_clip": 1.04858994,
      "balance_loss_mlp": 1.01726544,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.6455415000984257,
      "language_loss": 0.86513132,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88703996,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.578388214111328
    },
    {
      "auxiliary_loss_clip": 0.01159287,
      "auxiliary_loss_mlp": 0.01021154,
      "balance_loss_clip": 0.93558979,
      "balance_loss_mlp": 1.015131,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 2.0071032941095064,
      "language_loss": 0.69227475,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71407914,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.7359890937805176
    },
    {
      "auxiliary_loss_clip": 0.01164633,
      "auxiliary_loss_mlp": 0.01026874,
      "balance_loss_clip": 1.00997353,
      "balance_loss_mlp": 1.0200671,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 1.716656708408737,
      "language_loss": 0.82084852,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.8427636,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.6201748847961426
    },
    {
      "auxiliary_loss_clip": 0.01070838,
      "auxiliary_loss_mlp": 0.0111571,
      "balance_loss_clip": 0.89673948,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6650030674184362,
      "language_loss": 0.56354737,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58541286,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 4.148841142654419
    },
    {
      "auxiliary_loss_clip": 0.01162943,
      "auxiliary_loss_mlp": 0.01021881,
      "balance_loss_clip": 0.97018206,
      "balance_loss_mlp": 1.01436234,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 1.8685836508102796,
      "language_loss": 0.67414951,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.69599783,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 2.6589462757110596
    },
    {
      "auxiliary_loss_clip": 0.01167053,
      "auxiliary_loss_mlp": 0.01028933,
      "balance_loss_clip": 0.93172419,
      "balance_loss_mlp": 1.02163744,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.7120837005464438,
      "language_loss": 0.71530783,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.73726773,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.692288398742676
    },
    {
      "auxiliary_loss_clip": 0.01167857,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 0.97136462,
      "balance_loss_mlp": 1.0178771,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 2.769732800044904,
      "language_loss": 0.7501539,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77209342,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 3.561265707015991
    },
    {
      "auxiliary_loss_clip": 0.01163555,
      "auxiliary_loss_mlp": 0.01026848,
      "balance_loss_clip": 0.89181793,
      "balance_loss_mlp": 1.01901019,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 2.1650089821751513,
      "language_loss": 0.8041476,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82605159,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 3.647491931915283
    },
    {
      "auxiliary_loss_clip": 0.01166369,
      "auxiliary_loss_mlp": 0.01027246,
      "balance_loss_clip": 1.05022752,
      "balance_loss_mlp": 1.02020621,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 1.679815496281072,
      "language_loss": 0.80351621,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82545239,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.6415042877197266
    },
    {
      "auxiliary_loss_clip": 0.01169458,
      "auxiliary_loss_mlp": 0.01036476,
      "balance_loss_clip": 1.01270103,
      "balance_loss_mlp": 1.02891779,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 1.9775585577184498,
      "language_loss": 0.74924183,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77130121,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.612844228744507
    },
    {
      "auxiliary_loss_clip": 0.01172029,
      "auxiliary_loss_mlp": 0.01026118,
      "balance_loss_clip": 1.04835248,
      "balance_loss_mlp": 1.01870883,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 1.7476806112156285,
      "language_loss": 0.67028016,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.69226164,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.6073431968688965
    },
    {
      "auxiliary_loss_clip": 0.01169304,
      "auxiliary_loss_mlp": 0.01026926,
      "balance_loss_clip": 1.00890672,
      "balance_loss_mlp": 1.02040219,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.5633273721485916,
      "language_loss": 0.78271747,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80467975,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 2.6629106998443604
    },
    {
      "auxiliary_loss_clip": 0.01164429,
      "auxiliary_loss_mlp": 0.01027255,
      "balance_loss_clip": 0.96781647,
      "balance_loss_mlp": 1.02059174,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 1.887986886285108,
      "language_loss": 0.70781755,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.72973442,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.859995126724243
    },
    {
      "auxiliary_loss_clip": 0.01066735,
      "auxiliary_loss_mlp": 0.01001567,
      "balance_loss_clip": 0.85990167,
      "balance_loss_mlp": 0.9999221,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7480337288504864,
      "language_loss": 0.55478275,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.5754658,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 3.0857834815979004
    },
    {
      "auxiliary_loss_clip": 0.01063981,
      "auxiliary_loss_mlp": 0.0100233,
      "balance_loss_clip": 0.86215639,
      "balance_loss_mlp": 1.00075662,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.9252845387898442,
      "language_loss": 0.64802861,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66869175,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 3.236563205718994
    },
    {
      "auxiliary_loss_clip": 0.01165635,
      "auxiliary_loss_mlp": 0.01025132,
      "balance_loss_clip": 1.0488342,
      "balance_loss_mlp": 1.01826251,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 1.7977197060986831,
      "language_loss": 0.72622979,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.74813747,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 2.5328378677368164
    },
    {
      "auxiliary_loss_clip": 0.01170587,
      "auxiliary_loss_mlp": 0.01026339,
      "balance_loss_clip": 0.97194582,
      "balance_loss_mlp": 1.01933885,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 2.8088331455518247,
      "language_loss": 0.81466961,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83663893,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.6338915824890137
    },
    {
      "auxiliary_loss_clip": 0.01165083,
      "auxiliary_loss_mlp": 0.01027282,
      "balance_loss_clip": 0.97293377,
      "balance_loss_mlp": 1.01975441,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 1.6672464301706111,
      "language_loss": 0.72208643,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74401009,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 2.7007522583007812
    },
    {
      "auxiliary_loss_clip": 0.01147864,
      "auxiliary_loss_mlp": 0.01026925,
      "balance_loss_clip": 0.89230955,
      "balance_loss_mlp": 1.01930165,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 1.8092908026973484,
      "language_loss": 0.78263259,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80438054,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.740084648132324
    },
    {
      "auxiliary_loss_clip": 0.01169267,
      "auxiliary_loss_mlp": 0.01029785,
      "balance_loss_clip": 1.04860377,
      "balance_loss_mlp": 1.02194142,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 1.825806075991512,
      "language_loss": 0.74515188,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.76714236,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 2.6593308448791504
    },
    {
      "auxiliary_loss_clip": 0.0106956,
      "auxiliary_loss_mlp": 0.01115818,
      "balance_loss_clip": 0.86069429,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9145481839492435,
      "language_loss": 0.6639142,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.68576789,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.2651350498199463
    },
    {
      "auxiliary_loss_clip": 0.01163745,
      "auxiliary_loss_mlp": 0.01022655,
      "balance_loss_clip": 1.04669237,
      "balance_loss_mlp": 1.01597905,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 1.566569183642653,
      "language_loss": 0.7329694,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.7548334,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 2.6644320487976074
    },
    {
      "auxiliary_loss_clip": 0.01158949,
      "auxiliary_loss_mlp": 0.01023947,
      "balance_loss_clip": 0.93401229,
      "balance_loss_mlp": 1.01676512,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.5176636123333211,
      "language_loss": 0.74938244,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77121139,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.8888442516326904
    },
    {
      "auxiliary_loss_clip": 0.01166492,
      "auxiliary_loss_mlp": 0.01025355,
      "balance_loss_clip": 1.04812574,
      "balance_loss_mlp": 1.01840782,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 1.6289699660222339,
      "language_loss": 0.58568966,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60760814,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.798633575439453
    },
    {
      "auxiliary_loss_clip": 0.01164312,
      "auxiliary_loss_mlp": 0.01122836,
      "balance_loss_clip": 1.04660058,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 1.763310431574452,
      "language_loss": 0.71600735,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.73887885,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.607142210006714
    },
    {
      "auxiliary_loss_clip": 0.01167098,
      "auxiliary_loss_mlp": 0.01029803,
      "balance_loss_clip": 1.0482595,
      "balance_loss_mlp": 1.02319598,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 1.8370631454429338,
      "language_loss": 0.53867388,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.56064284,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 3.451992988586426
    },
    {
      "auxiliary_loss_clip": 0.01166899,
      "auxiliary_loss_mlp": 0.01029519,
      "balance_loss_clip": 1.01026487,
      "balance_loss_mlp": 1.02271867,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.4701086914199057,
      "language_loss": 0.65427738,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.67624152,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.9009690284729004
    },
    {
      "auxiliary_loss_clip": 0.01164709,
      "auxiliary_loss_mlp": 0.01022891,
      "balance_loss_clip": 1.00780916,
      "balance_loss_mlp": 1.01638508,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 1.8594759175357265,
      "language_loss": 0.75489485,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.77677083,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.567470073699951
    },
    {
      "auxiliary_loss_clip": 0.01163671,
      "auxiliary_loss_mlp": 0.01025071,
      "balance_loss_clip": 1.00978756,
      "balance_loss_mlp": 1.0178349,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 2.490762461703964,
      "language_loss": 0.83923918,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.86112654,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.6674814224243164
    },
    {
      "auxiliary_loss_clip": 0.01063372,
      "auxiliary_loss_mlp": 0.01002135,
      "balance_loss_clip": 0.97183841,
      "balance_loss_mlp": 1.00053787,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.7686254594526672,
      "language_loss": 0.56823051,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58888561,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 4.06515908241272
    },
    {
      "auxiliary_loss_clip": 0.01164167,
      "auxiliary_loss_mlp": 0.01023747,
      "balance_loss_clip": 0.97086668,
      "balance_loss_mlp": 1.01654708,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 1.8747564602522218,
      "language_loss": 0.62728262,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.64916176,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 2.6426708698272705
    },
    {
      "auxiliary_loss_clip": 0.01164236,
      "auxiliary_loss_mlp": 0.0103361,
      "balance_loss_clip": 0.92941672,
      "balance_loss_mlp": 1.02640963,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 1.8931096557462974,
      "language_loss": 0.76709664,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.78907514,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.682583808898926
    },
    {
      "auxiliary_loss_clip": 0.01163186,
      "auxiliary_loss_mlp": 0.01023828,
      "balance_loss_clip": 1.04866278,
      "balance_loss_mlp": 1.01743817,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 1.4977946885192988,
      "language_loss": 0.7972517,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.81912178,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 3.470402956008911
    },
    {
      "auxiliary_loss_clip": 0.01062787,
      "auxiliary_loss_mlp": 0.01001637,
      "balance_loss_clip": 0.97124612,
      "balance_loss_mlp": 0.99996775,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8149062835577128,
      "language_loss": 0.63982129,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.66046554,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 4.1399595737457275
    },
    {
      "auxiliary_loss_clip": 0.01163749,
      "auxiliary_loss_mlp": 0.01022272,
      "balance_loss_clip": 1.01067543,
      "balance_loss_mlp": 1.01514924,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.313800957400373,
      "language_loss": 0.75035578,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.77221596,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 2.6512868404388428
    },
    {
      "auxiliary_loss_clip": 0.01165148,
      "auxiliary_loss_mlp": 0.01020448,
      "balance_loss_clip": 1.04774618,
      "balance_loss_mlp": 1.01384044,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 3.220475189946779,
      "language_loss": 0.76249242,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78434837,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.5723791122436523
    },
    {
      "auxiliary_loss_clip": 0.01065682,
      "auxiliary_loss_mlp": 0.01000896,
      "balance_loss_clip": 0.82452846,
      "balance_loss_mlp": 0.99933398,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 1.13152417204504,
      "language_loss": 0.62326765,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64393342,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.4309065341949463
    },
    {
      "auxiliary_loss_clip": 0.01060378,
      "auxiliary_loss_mlp": 0.01001432,
      "balance_loss_clip": 0.97238255,
      "balance_loss_mlp": 0.99988234,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9315963032920705,
      "language_loss": 0.57866561,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.5992837,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.3943445682525635
    },
    {
      "auxiliary_loss_clip": 0.01164499,
      "auxiliary_loss_mlp": 0.01026844,
      "balance_loss_clip": 1.04596472,
      "balance_loss_mlp": 1.02001047,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 1.6746379547653911,
      "language_loss": 0.76014113,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78205454,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.6763916015625
    },
    {
      "auxiliary_loss_clip": 0.01170974,
      "auxiliary_loss_mlp": 0.0102778,
      "balance_loss_clip": 1.04936361,
      "balance_loss_mlp": 1.02109814,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 4.490137689617844,
      "language_loss": 0.73160934,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.7535969,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.5964298248291016
    },
    {
      "auxiliary_loss_clip": 0.01164163,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.0116564,
      "balance_loss_mlp": 1.02186775,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 2.3853480502777114,
      "language_loss": 0.66555536,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.68748426,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.6246073246002197
    },
    {
      "auxiliary_loss_clip": 0.011455,
      "auxiliary_loss_mlp": 0.01027752,
      "balance_loss_clip": 0.96563566,
      "balance_loss_mlp": 1.0207485,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 1.587776593257799,
      "language_loss": 0.72415131,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74588388,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 2.729092597961426
    },
    {
      "auxiliary_loss_clip": 0.01165671,
      "auxiliary_loss_mlp": 0.01025538,
      "balance_loss_clip": 1.012398,
      "balance_loss_mlp": 1.01798606,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 5.576795259408118,
      "language_loss": 0.79582298,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.81773502,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.578615427017212
    },
    {
      "auxiliary_loss_clip": 0.0116096,
      "auxiliary_loss_mlp": 0.01027122,
      "balance_loss_clip": 0.9332608,
      "balance_loss_mlp": 1.02020526,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 1.7608110758927256,
      "language_loss": 0.74199665,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76387751,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 2.7559070587158203
    },
    {
      "auxiliary_loss_clip": 0.01150118,
      "auxiliary_loss_mlp": 0.01029194,
      "balance_loss_clip": 0.96741658,
      "balance_loss_mlp": 1.02213693,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 1.7928345718867456,
      "language_loss": 0.85180163,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87359476,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 2.597360849380493
    },
    {
      "auxiliary_loss_clip": 0.0116571,
      "auxiliary_loss_mlp": 0.01020522,
      "balance_loss_clip": 0.97277069,
      "balance_loss_mlp": 1.01434684,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.6940952795190216,
      "language_loss": 0.81192613,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.83378845,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.6592462062835693
    },
    {
      "auxiliary_loss_clip": 0.01164649,
      "auxiliary_loss_mlp": 0.01026922,
      "balance_loss_clip": 0.97263277,
      "balance_loss_mlp": 1.01909637,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 1.8443905491302748,
      "language_loss": 0.74378705,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.76570278,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 2.6229827404022217
    },
    {
      "auxiliary_loss_clip": 0.01062387,
      "auxiliary_loss_mlp": 0.00999472,
      "balance_loss_clip": 0.93608725,
      "balance_loss_mlp": 0.99798203,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 2.0582119719908905,
      "language_loss": 0.59123123,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61184978,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.2879021167755127
    },
    {
      "auxiliary_loss_clip": 0.01166659,
      "auxiliary_loss_mlp": 0.01027504,
      "balance_loss_clip": 1.01058877,
      "balance_loss_mlp": 1.0200479,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 2.0726051808935497,
      "language_loss": 0.79156566,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.81350732,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.657559633255005
    },
    {
      "auxiliary_loss_clip": 0.01170843,
      "auxiliary_loss_mlp": 0.01024898,
      "balance_loss_clip": 1.01012814,
      "balance_loss_mlp": 1.01767373,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 1.8537462313035031,
      "language_loss": 0.76565611,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.78761351,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.656602621078491
    },
    {
      "auxiliary_loss_clip": 0.01166929,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 0.85927773,
      "balance_loss_mlp": 1.02095342,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 2.8743783543214505,
      "language_loss": 0.64031518,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66227305,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.757716417312622
    },
    {
      "auxiliary_loss_clip": 0.01168862,
      "auxiliary_loss_mlp": 0.01026478,
      "balance_loss_clip": 1.01075733,
      "balance_loss_mlp": 1.01939678,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 2.560201045854158,
      "language_loss": 0.8300274,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.85198081,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 3.51556396484375
    },
    {
      "auxiliary_loss_clip": 0.01168745,
      "auxiliary_loss_mlp": 0.0102213,
      "balance_loss_clip": 1.04686403,
      "balance_loss_mlp": 1.01493001,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 1.8066455056193642,
      "language_loss": 0.77228224,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.794191,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.5792782306671143
    },
    {
      "auxiliary_loss_clip": 0.01167759,
      "auxiliary_loss_mlp": 0.01027944,
      "balance_loss_clip": 1.04687452,
      "balance_loss_mlp": 1.020437,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 2.6573867773225146,
      "language_loss": 0.67354417,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69550121,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.617441415786743
    },
    {
      "auxiliary_loss_clip": 0.01172172,
      "auxiliary_loss_mlp": 0.01025422,
      "balance_loss_clip": 0.93478703,
      "balance_loss_mlp": 1.01827812,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.078038105416487,
      "language_loss": 0.65826756,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.68024349,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.7131454944610596
    },
    {
      "auxiliary_loss_clip": 0.01166899,
      "auxiliary_loss_mlp": 0.01026932,
      "balance_loss_clip": 1.04714155,
      "balance_loss_mlp": 1.02016687,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 1.551349725105174,
      "language_loss": 0.79165417,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81359243,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 3.5046935081481934
    },
    {
      "auxiliary_loss_clip": 0.01164598,
      "auxiliary_loss_mlp": 0.0102762,
      "balance_loss_clip": 1.00952697,
      "balance_loss_mlp": 1.02089655,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.2730894742568544,
      "language_loss": 0.73027259,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75219476,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 2.558929920196533
    },
    {
      "auxiliary_loss_clip": 0.01156368,
      "auxiliary_loss_mlp": 0.01024816,
      "balance_loss_clip": 0.89169008,
      "balance_loss_mlp": 1.01819372,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 2.0851816834969266,
      "language_loss": 0.77602726,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.79783905,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.730067491531372
    },
    {
      "auxiliary_loss_clip": 0.01162971,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.01063073,
      "balance_loss_mlp": 1.02033639,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 1.7104776409032523,
      "language_loss": 0.90219837,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92409444,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 3.9098713397979736
    },
    {
      "auxiliary_loss_clip": 0.01163049,
      "auxiliary_loss_mlp": 0.01026939,
      "balance_loss_clip": 1.04590726,
      "balance_loss_mlp": 1.02000451,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 5.7599584421087,
      "language_loss": 0.80090749,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.82280737,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 3.5852839946746826
    },
    {
      "auxiliary_loss_clip": 0.01164977,
      "auxiliary_loss_mlp": 0.01022878,
      "balance_loss_clip": 1.01027036,
      "balance_loss_mlp": 1.01624084,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 1.9687119967511213,
      "language_loss": 0.78093457,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80281311,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.6179163455963135
    },
    {
      "auxiliary_loss_clip": 0.01156476,
      "auxiliary_loss_mlp": 0.01123023,
      "balance_loss_clip": 1.00437474,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 1.831885191217578,
      "language_loss": 0.66598797,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.68878299,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.615046501159668
    },
    {
      "auxiliary_loss_clip": 0.0116586,
      "auxiliary_loss_mlp": 0.01024957,
      "balance_loss_clip": 0.96847975,
      "balance_loss_mlp": 1.0182724,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.7555675351769455,
      "language_loss": 0.88738728,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.90929544,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 2.769052743911743
    },
    {
      "auxiliary_loss_clip": 0.01162516,
      "auxiliary_loss_mlp": 0.01025708,
      "balance_loss_clip": 1.00980568,
      "balance_loss_mlp": 1.01898468,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 1.5152389361756418,
      "language_loss": 0.78895694,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.81083918,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.63641357421875
    },
    {
      "auxiliary_loss_clip": 0.01169719,
      "auxiliary_loss_mlp": 0.01122301,
      "balance_loss_clip": 1.0158776,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 2.3522982010476503,
      "language_loss": 0.85887456,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.88179469,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.6393754482269287
    },
    {
      "auxiliary_loss_clip": 0.01159525,
      "auxiliary_loss_mlp": 0.01023213,
      "balance_loss_clip": 0.96972847,
      "balance_loss_mlp": 1.01659131,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 1.6155605263294175,
      "language_loss": 0.87227589,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89410329,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.5866596698760986
    },
    {
      "auxiliary_loss_clip": 0.01171407,
      "auxiliary_loss_mlp": 0.01034737,
      "balance_loss_clip": 0.93224323,
      "balance_loss_mlp": 1.02685118,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 2.5435122611309096,
      "language_loss": 0.77412152,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.79618299,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.6365954875946045
    },
    {
      "auxiliary_loss_clip": 0.01164935,
      "auxiliary_loss_mlp": 0.01022083,
      "balance_loss_clip": 0.97051859,
      "balance_loss_mlp": 1.01544046,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 1.799747766849792,
      "language_loss": 0.85147786,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87334806,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 2.612063407897949
    },
    {
      "auxiliary_loss_clip": 0.01168991,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.04967237,
      "balance_loss_mlp": 1.01942968,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 2.2842580341695853,
      "language_loss": 0.86632562,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.88827848,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.5803277492523193
    },
    {
      "auxiliary_loss_clip": 0.01173051,
      "auxiliary_loss_mlp": 0.01033085,
      "balance_loss_clip": 0.93296456,
      "balance_loss_mlp": 1.02585161,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 1.8626401786830256,
      "language_loss": 0.66643202,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.68849337,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 2.648569345474243
    },
    {
      "auxiliary_loss_clip": 0.01154619,
      "auxiliary_loss_mlp": 0.01030703,
      "balance_loss_clip": 0.96678007,
      "balance_loss_mlp": 1.02358663,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 1.8508161967309873,
      "language_loss": 0.68553066,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.70738387,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 2.608348846435547
    },
    {
      "auxiliary_loss_clip": 0.01165668,
      "auxiliary_loss_mlp": 0.01022493,
      "balance_loss_clip": 0.93417025,
      "balance_loss_mlp": 1.01536977,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 1.6975032106181815,
      "language_loss": 0.7910918,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.8129735,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 2.7241203784942627
    },
    {
      "auxiliary_loss_clip": 0.01162891,
      "auxiliary_loss_mlp": 0.01021366,
      "balance_loss_clip": 0.9742012,
      "balance_loss_mlp": 1.01481819,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 2.2204259994331754,
      "language_loss": 0.84665453,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.86849713,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.633740186691284
    },
    {
      "auxiliary_loss_clip": 0.01161513,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 0.97028232,
      "balance_loss_mlp": 1.02212644,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 1.6268926764126765,
      "language_loss": 0.74858922,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77050221,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 2.6707959175109863
    },
    {
      "auxiliary_loss_clip": 0.01161783,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 0.89213616,
      "balance_loss_mlp": 1.01876092,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 1.647914577968033,
      "language_loss": 0.72030962,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.74218845,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.744957447052002
    },
    {
      "auxiliary_loss_clip": 0.01073478,
      "auxiliary_loss_mlp": 0.01000956,
      "balance_loss_clip": 0.82451499,
      "balance_loss_mlp": 0.99931043,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.6931542673078303,
      "language_loss": 0.56654179,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.58728617,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 3.524487018585205
    },
    {
      "auxiliary_loss_clip": 0.01062845,
      "auxiliary_loss_mlp": 0.0100206,
      "balance_loss_clip": 0.97158283,
      "balance_loss_mlp": 1.00042713,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7686797766620697,
      "language_loss": 0.54722321,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.56787229,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.412020444869995
    },
    {
      "auxiliary_loss_clip": 0.01157313,
      "auxiliary_loss_mlp": 0.01022782,
      "balance_loss_clip": 0.89678133,
      "balance_loss_mlp": 1.01508713,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 2.395323250200306,
      "language_loss": 0.79616642,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.81796736,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 3.573500394821167
    },
    {
      "auxiliary_loss_clip": 0.01170147,
      "auxiliary_loss_mlp": 0.01122557,
      "balance_loss_clip": 1.01114511,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 1.82869259290222,
      "language_loss": 0.74276221,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.76568919,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 2.748293399810791
    },
    {
      "auxiliary_loss_clip": 0.01165232,
      "auxiliary_loss_mlp": 0.01026429,
      "balance_loss_clip": 0.93653607,
      "balance_loss_mlp": 1.01946449,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 1.8512555567284814,
      "language_loss": 0.83416808,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.8560847,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.706824779510498
    },
    {
      "auxiliary_loss_clip": 0.01169536,
      "auxiliary_loss_mlp": 0.01122933,
      "balance_loss_clip": 1.0080899,
      "balance_loss_mlp": 0.0,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 2.1824391267443795,
      "language_loss": 0.84302038,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.8659451,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.643951177597046
    },
    {
      "auxiliary_loss_clip": 0.01073371,
      "auxiliary_loss_mlp": 0.01001412,
      "balance_loss_clip": 0.98598933,
      "balance_loss_mlp": 0.99979037,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.8569559107436281,
      "language_loss": 0.56484151,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.58558929,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.2400670051574707
    },
    {
      "auxiliary_loss_clip": 0.0116022,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 0.96959984,
      "balance_loss_mlp": 1.01999247,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.0880748428346414,
      "language_loss": 0.78242505,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.8043012,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 3.600187301635742
    },
    {
      "auxiliary_loss_clip": 0.01160049,
      "auxiliary_loss_mlp": 0.01026294,
      "balance_loss_clip": 0.97172487,
      "balance_loss_mlp": 1.01909685,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 2.053491124720334,
      "language_loss": 0.72320122,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74506468,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.6170742511749268
    },
    {
      "auxiliary_loss_clip": 0.01163176,
      "auxiliary_loss_mlp": 0.01021231,
      "balance_loss_clip": 1.01040351,
      "balance_loss_mlp": 1.01473427,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 1.812978620591092,
      "language_loss": 0.68013465,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70197874,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.744903802871704
    },
    {
      "auxiliary_loss_clip": 0.01164746,
      "auxiliary_loss_mlp": 0.01021323,
      "balance_loss_clip": 1.0457902,
      "balance_loss_mlp": 1.01433432,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 2.0042195787716914,
      "language_loss": 0.70611674,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.7279774,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 4.610821485519409
    },
    {
      "auxiliary_loss_clip": 0.0115707,
      "auxiliary_loss_mlp": 0.01025365,
      "balance_loss_clip": 0.93226588,
      "balance_loss_mlp": 1.0190351,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 1.5049999964645169,
      "language_loss": 0.74030733,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.76213169,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.7123286724090576
    },
    {
      "auxiliary_loss_clip": 0.01058191,
      "auxiliary_loss_mlp": 0.01001896,
      "balance_loss_clip": 0.97192729,
      "balance_loss_mlp": 1.00035846,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7630525208965675,
      "language_loss": 0.53009862,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.55069947,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.1590769290924072
    },
    {
      "auxiliary_loss_clip": 0.01160703,
      "auxiliary_loss_mlp": 0.01030327,
      "balance_loss_clip": 0.89432895,
      "balance_loss_mlp": 1.02310038,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 1.9606203493051924,
      "language_loss": 0.70945072,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.73136103,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 2.6654739379882812
    },
    {
      "auxiliary_loss_clip": 0.01166043,
      "auxiliary_loss_mlp": 0.0102464,
      "balance_loss_clip": 0.97206795,
      "balance_loss_mlp": 1.01788986,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 1.8598724207505481,
      "language_loss": 0.73076415,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.75267094,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 2.645030975341797
    },
    {
      "auxiliary_loss_clip": 0.01160465,
      "auxiliary_loss_mlp": 0.0103052,
      "balance_loss_clip": 0.9659071,
      "balance_loss_mlp": 1.0235405,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 1.8711134544433823,
      "language_loss": 0.73867983,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76058972,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.672119140625
    },
    {
      "auxiliary_loss_clip": 0.01164986,
      "auxiliary_loss_mlp": 0.01026686,
      "balance_loss_clip": 0.96966934,
      "balance_loss_mlp": 1.01989985,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.4485364535688912,
      "language_loss": 0.79243016,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81434685,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.6965179443359375
    },
    {
      "auxiliary_loss_clip": 0.01170953,
      "auxiliary_loss_mlp": 0.01123003,
      "balance_loss_clip": 1.00967932,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 1.7113021528330352,
      "language_loss": 0.66355354,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.68649304,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 2.625295400619507
    },
    {
      "auxiliary_loss_clip": 0.01162318,
      "auxiliary_loss_mlp": 0.01027364,
      "balance_loss_clip": 0.96785259,
      "balance_loss_mlp": 1.02025914,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 2.0427518289258013,
      "language_loss": 0.78211856,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.8040154,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 2.679164171218872
    },
    {
      "auxiliary_loss_clip": 0.01163345,
      "auxiliary_loss_mlp": 0.01021599,
      "balance_loss_clip": 1.0095439,
      "balance_loss_mlp": 1.01471186,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 2.570557298651503,
      "language_loss": 0.75913262,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78098208,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.6197729110717773
    },
    {
      "auxiliary_loss_clip": 0.01166415,
      "auxiliary_loss_mlp": 0.0102418,
      "balance_loss_clip": 1.01116228,
      "balance_loss_mlp": 1.01731646,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 1.957359914766101,
      "language_loss": 0.70300716,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.72491312,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 3.002825975418091
    },
    {
      "auxiliary_loss_clip": 0.01164942,
      "auxiliary_loss_mlp": 0.01025631,
      "balance_loss_clip": 0.96890211,
      "balance_loss_mlp": 1.01821315,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 1.9691379054888773,
      "language_loss": 0.76065999,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78256577,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 2.677988290786743
    },
    {
      "auxiliary_loss_clip": 0.01168197,
      "auxiliary_loss_mlp": 0.01026095,
      "balance_loss_clip": 1.04852986,
      "balance_loss_mlp": 1.01885259,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.2664716389665696,
      "language_loss": 0.80383611,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82577902,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.5361454486846924
    },
    {
      "auxiliary_loss_clip": 0.01169045,
      "auxiliary_loss_mlp": 0.01023474,
      "balance_loss_clip": 0.9715482,
      "balance_loss_mlp": 1.01682806,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.266982137985291,
      "language_loss": 0.61744297,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.63936818,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 2.6838104724884033
    },
    {
      "auxiliary_loss_clip": 0.01167363,
      "auxiliary_loss_mlp": 0.01022919,
      "balance_loss_clip": 0.89214635,
      "balance_loss_mlp": 1.01535487,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 2.10133825344804,
      "language_loss": 0.7442314,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76613426,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.7241647243499756
    },
    {
      "auxiliary_loss_clip": 0.01165629,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.04669356,
      "balance_loss_mlp": 1.02079618,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 2.3270295261386247,
      "language_loss": 0.78946996,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81139976,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.5623114109039307
    },
    {
      "auxiliary_loss_clip": 0.01163609,
      "auxiliary_loss_mlp": 0.01022803,
      "balance_loss_clip": 1.04551482,
      "balance_loss_mlp": 1.01589525,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 3.0146839743685097,
      "language_loss": 0.79528117,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.81714529,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.646259307861328
    },
    {
      "auxiliary_loss_clip": 0.01171904,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 0.97464943,
      "balance_loss_mlp": 1.01952696,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.1478608907573333,
      "language_loss": 0.72234541,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.74433398,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 2.6136178970336914
    },
    {
      "auxiliary_loss_clip": 0.01167113,
      "auxiliary_loss_mlp": 0.01027033,
      "balance_loss_clip": 1.00881362,
      "balance_loss_mlp": 1.01940632,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 1.8208457879531508,
      "language_loss": 0.78419662,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80613804,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 3.3915750980377197
    },
    {
      "auxiliary_loss_clip": 0.01168499,
      "auxiliary_loss_mlp": 0.01027663,
      "balance_loss_clip": 1.01158905,
      "balance_loss_mlp": 1.02054358,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 2.1468185766946495,
      "language_loss": 0.71544302,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.73740464,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.5973691940307617
    },
    {
      "auxiliary_loss_clip": 0.01159595,
      "auxiliary_loss_mlp": 0.0102463,
      "balance_loss_clip": 0.97081137,
      "balance_loss_mlp": 1.01722109,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 1.7305912020972654,
      "language_loss": 0.6540525,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.67589474,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.6994831562042236
    },
    {
      "auxiliary_loss_clip": 0.01163825,
      "auxiliary_loss_mlp": 0.01028718,
      "balance_loss_clip": 1.00849986,
      "balance_loss_mlp": 1.02185464,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 1.6713707991762699,
      "language_loss": 0.76938224,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.79130769,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 2.6962170600891113
    },
    {
      "auxiliary_loss_clip": 0.01156887,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.00869107,
      "balance_loss_mlp": 1.01873732,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 1.711370649030336,
      "language_loss": 0.72094798,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.7427749,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.6512696743011475
    },
    {
      "auxiliary_loss_clip": 0.01153024,
      "auxiliary_loss_mlp": 0.01028944,
      "balance_loss_clip": 0.85716784,
      "balance_loss_mlp": 1.02153492,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 1.922795423468803,
      "language_loss": 0.8172965,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.83911622,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 3.726722478866577
    },
    {
      "auxiliary_loss_clip": 0.0116976,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 0.97542858,
      "balance_loss_mlp": 1.01952434,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.6819253308771485,
      "language_loss": 0.82078791,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84275264,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.7254974842071533
    },
    {
      "auxiliary_loss_clip": 0.01168904,
      "auxiliary_loss_mlp": 0.01025711,
      "balance_loss_clip": 0.93550456,
      "balance_loss_mlp": 1.01868963,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 1.7172103641970446,
      "language_loss": 0.78758228,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80952835,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.6196625232696533
    },
    {
      "auxiliary_loss_clip": 0.0116861,
      "auxiliary_loss_mlp": 0.0102364,
      "balance_loss_clip": 1.04730618,
      "balance_loss_mlp": 1.0162375,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 1.8913174482146013,
      "language_loss": 0.79261577,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.8145383,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 4.408765554428101
    },
    {
      "auxiliary_loss_clip": 0.01151151,
      "auxiliary_loss_mlp": 0.01025786,
      "balance_loss_clip": 0.92904705,
      "balance_loss_mlp": 1.01899648,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.698396961687031,
      "language_loss": 0.69828266,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.720052,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.84976863861084
    },
    {
      "auxiliary_loss_clip": 0.01165256,
      "auxiliary_loss_mlp": 0.01023459,
      "balance_loss_clip": 0.97265953,
      "balance_loss_mlp": 1.01634216,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 2.0897619535602625,
      "language_loss": 0.75965303,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78154016,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.801532745361328
    },
    {
      "auxiliary_loss_clip": 0.01168525,
      "auxiliary_loss_mlp": 0.01023065,
      "balance_loss_clip": 0.97058594,
      "balance_loss_mlp": 1.01666033,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 1.9399835543746753,
      "language_loss": 0.78113067,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.80304658,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 2.663590908050537
    },
    {
      "auxiliary_loss_clip": 0.01161898,
      "auxiliary_loss_mlp": 0.01023307,
      "balance_loss_clip": 0.97322178,
      "balance_loss_mlp": 1.01665199,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 1.8492723960334723,
      "language_loss": 0.72298706,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74483913,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 2.6704955101013184
    },
    {
      "auxiliary_loss_clip": 0.01169867,
      "auxiliary_loss_mlp": 0.0102019,
      "balance_loss_clip": 0.97146213,
      "balance_loss_mlp": 1.01355636,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 1.637801906845894,
      "language_loss": 0.74152547,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.76342607,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.701488733291626
    },
    {
      "auxiliary_loss_clip": 0.01151431,
      "auxiliary_loss_mlp": 0.01025319,
      "balance_loss_clip": 0.88937199,
      "balance_loss_mlp": 1.01823199,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.5027925418667094,
      "language_loss": 0.85636663,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.87813413,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.8829166889190674
    },
    {
      "auxiliary_loss_clip": 0.01160751,
      "auxiliary_loss_mlp": 0.01021361,
      "balance_loss_clip": 1.0090816,
      "balance_loss_mlp": 1.01478994,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 2.247122411105706,
      "language_loss": 0.70909047,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.73091161,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 2.711308240890503
    },
    {
      "auxiliary_loss_clip": 0.01154437,
      "auxiliary_loss_mlp": 0.01030166,
      "balance_loss_clip": 1.00690317,
      "balance_loss_mlp": 1.02309966,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 1.5137661012661088,
      "language_loss": 0.78860199,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81044805,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.685457944869995
    },
    {
      "auxiliary_loss_clip": 0.01166893,
      "auxiliary_loss_mlp": 0.01031068,
      "balance_loss_clip": 0.97021288,
      "balance_loss_mlp": 1.02432036,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 1.597372932829824,
      "language_loss": 0.74777013,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.76974976,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.6743557453155518
    },
    {
      "auxiliary_loss_clip": 0.01171657,
      "auxiliary_loss_mlp": 0.01027314,
      "balance_loss_clip": 0.9335472,
      "balance_loss_mlp": 1.02080202,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 1.9394967876353773,
      "language_loss": 0.75719965,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.77918935,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 2.6516969203948975
    },
    {
      "auxiliary_loss_clip": 0.01171633,
      "auxiliary_loss_mlp": 0.01023312,
      "balance_loss_clip": 0.93296921,
      "balance_loss_mlp": 1.01683652,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 2.62775056793039,
      "language_loss": 0.81519181,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83714128,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 2.6607816219329834
    },
    {
      "auxiliary_loss_clip": 0.01151329,
      "auxiliary_loss_mlp": 0.01027494,
      "balance_loss_clip": 0.92861384,
      "balance_loss_mlp": 1.02048481,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 2.306074368056061,
      "language_loss": 0.67680478,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.69859302,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 2.681562900543213
    },
    {
      "auxiliary_loss_clip": 0.0116511,
      "auxiliary_loss_mlp": 0.01024283,
      "balance_loss_clip": 0.97200632,
      "balance_loss_mlp": 1.01682925,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 1.7205988704405044,
      "language_loss": 0.66663766,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.68853164,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 2.715153932571411
    },
    {
      "auxiliary_loss_clip": 0.01167907,
      "auxiliary_loss_mlp": 0.01021522,
      "balance_loss_clip": 1.01129913,
      "balance_loss_mlp": 1.01434875,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 1.7916501810029941,
      "language_loss": 0.74953103,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77142531,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.5750722885131836
    },
    {
      "auxiliary_loss_clip": 0.01172192,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 0.8191216,
      "balance_loss_mlp": 1.02207065,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 2.932080434468456,
      "language_loss": 0.75165218,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77366853,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.760169267654419
    },
    {
      "auxiliary_loss_clip": 0.01160949,
      "auxiliary_loss_mlp": 0.01023787,
      "balance_loss_clip": 1.00942051,
      "balance_loss_mlp": 1.01672673,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 1.9065565431500002,
      "language_loss": 0.83366573,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.8555131,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 2.772260904312134
    },
    {
      "auxiliary_loss_clip": 0.01160368,
      "auxiliary_loss_mlp": 0.01023676,
      "balance_loss_clip": 0.9681856,
      "balance_loss_mlp": 1.01626682,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 1.7245259128839132,
      "language_loss": 0.61667854,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.63851899,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.6019158363342285
    },
    {
      "auxiliary_loss_clip": 0.01065144,
      "auxiliary_loss_mlp": 0.01000036,
      "balance_loss_clip": 0.93391061,
      "balance_loss_mlp": 0.9983905,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.7565232392404639,
      "language_loss": 0.57128,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59193182,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 4.146923065185547
    },
    {
      "auxiliary_loss_clip": 0.0116645,
      "auxiliary_loss_mlp": 0.01029573,
      "balance_loss_clip": 1.01061678,
      "balance_loss_mlp": 1.02196193,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 3.0188668113610673,
      "language_loss": 0.78229821,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.80425841,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.7023537158966064
    },
    {
      "auxiliary_loss_clip": 0.0106325,
      "auxiliary_loss_mlp": 0.01002471,
      "balance_loss_clip": 0.97105187,
      "balance_loss_mlp": 1.00074244,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.8011484214265037,
      "language_loss": 0.54337788,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56403512,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.2080705165863037
    },
    {
      "auxiliary_loss_clip": 0.01159955,
      "auxiliary_loss_mlp": 0.01027352,
      "balance_loss_clip": 0.89508832,
      "balance_loss_mlp": 1.019943,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.7056819186991703,
      "language_loss": 0.7727145,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79458755,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 2.749969005584717
    },
    {
      "auxiliary_loss_clip": 0.01166625,
      "auxiliary_loss_mlp": 0.01026214,
      "balance_loss_clip": 1.04614031,
      "balance_loss_mlp": 1.01841724,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.2030623525628803,
      "language_loss": 0.59820843,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.6201368,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.599276542663574
    },
    {
      "auxiliary_loss_clip": 0.0116031,
      "auxiliary_loss_mlp": 0.01024437,
      "balance_loss_clip": 1.00779307,
      "balance_loss_mlp": 1.01771951,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 1.7922036757595903,
      "language_loss": 0.74630737,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.76815486,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 3.7544806003570557
    },
    {
      "auxiliary_loss_clip": 0.01156669,
      "auxiliary_loss_mlp": 0.0102609,
      "balance_loss_clip": 0.93267667,
      "balance_loss_mlp": 1.01892877,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 2.46075534952158,
      "language_loss": 0.76488352,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.7867111,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.679151773452759
    },
    {
      "auxiliary_loss_clip": 0.01164664,
      "auxiliary_loss_mlp": 0.01122289,
      "balance_loss_clip": 0.93277287,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 1.7702272233034952,
      "language_loss": 0.74732268,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.77019215,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.66875958442688
    },
    {
      "auxiliary_loss_clip": 0.01165654,
      "auxiliary_loss_mlp": 0.01033866,
      "balance_loss_clip": 0.93103397,
      "balance_loss_mlp": 1.02656722,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 1.62548125202387,
      "language_loss": 0.72444129,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.74643654,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 3.5956056118011475
    },
    {
      "auxiliary_loss_clip": 0.01163395,
      "auxiliary_loss_mlp": 0.01029994,
      "balance_loss_clip": 0.96772647,
      "balance_loss_mlp": 1.02198887,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.6788841730503468,
      "language_loss": 0.76965332,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.79158717,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 3.5570380687713623
    },
    {
      "auxiliary_loss_clip": 0.01156624,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 0.92935854,
      "balance_loss_mlp": 1.01845717,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 1.5906432203935992,
      "language_loss": 0.59158313,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61340153,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 2.7273998260498047
    },
    {
      "auxiliary_loss_clip": 0.01149783,
      "auxiliary_loss_mlp": 0.01021261,
      "balance_loss_clip": 0.92686409,
      "balance_loss_mlp": 1.01426673,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 2.2538066191169417,
      "language_loss": 0.72995245,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.75166297,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 2.7869362831115723
    },
    {
      "auxiliary_loss_clip": 0.01163685,
      "auxiliary_loss_mlp": 0.01023974,
      "balance_loss_clip": 1.00689626,
      "balance_loss_mlp": 1.01669002,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.5617971062777107,
      "language_loss": 0.66179192,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.68366849,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.6545679569244385
    },
    {
      "auxiliary_loss_clip": 0.01162159,
      "auxiliary_loss_mlp": 0.01024843,
      "balance_loss_clip": 1.00844502,
      "balance_loss_mlp": 1.01770806,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 2.1252057929700072,
      "language_loss": 0.81085718,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.83272719,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.659759044647217
    },
    {
      "auxiliary_loss_clip": 0.01168393,
      "auxiliary_loss_mlp": 0.0102602,
      "balance_loss_clip": 1.01186252,
      "balance_loss_mlp": 1.0184207,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 3.055727072129161,
      "language_loss": 0.71032351,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.73226762,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 2.6239116191864014
    },
    {
      "auxiliary_loss_clip": 0.01168769,
      "auxiliary_loss_mlp": 0.01022623,
      "balance_loss_clip": 1.04737115,
      "balance_loss_mlp": 1.01545227,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 1.5816829918795272,
      "language_loss": 0.79314965,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.8150636,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.6242599487304688
    },
    {
      "auxiliary_loss_clip": 0.0116747,
      "auxiliary_loss_mlp": 0.01025731,
      "balance_loss_clip": 0.97322142,
      "balance_loss_mlp": 1.01817954,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 1.922816723987633,
      "language_loss": 0.63917375,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66110575,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 2.7711844444274902
    },
    {
      "auxiliary_loss_clip": 0.01168531,
      "auxiliary_loss_mlp": 0.01024693,
      "balance_loss_clip": 1.04853284,
      "balance_loss_mlp": 1.01803541,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.6190943305526697,
      "language_loss": 0.71404827,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.73598051,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.5789401531219482
    },
    {
      "auxiliary_loss_clip": 0.01165101,
      "auxiliary_loss_mlp": 0.01032369,
      "balance_loss_clip": 0.97101498,
      "balance_loss_mlp": 1.02565742,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 3.3984480152656427,
      "language_loss": 0.77121544,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.79319018,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 2.6363699436187744
    },
    {
      "auxiliary_loss_clip": 0.01154271,
      "auxiliary_loss_mlp": 0.01017695,
      "balance_loss_clip": 0.96802449,
      "balance_loss_mlp": 1.01049733,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 1.6782689573500125,
      "language_loss": 0.71574515,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.73746485,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.700345993041992
    },
    {
      "auxiliary_loss_clip": 0.01152499,
      "auxiliary_loss_mlp": 0.0102755,
      "balance_loss_clip": 0.92811888,
      "balance_loss_mlp": 1.02065396,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.589665159905405,
      "language_loss": 0.77457529,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79637575,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 2.7193312644958496
    },
    {
      "auxiliary_loss_clip": 0.01170456,
      "auxiliary_loss_mlp": 0.01122043,
      "balance_loss_clip": 1.01083624,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.672298849094127,
      "language_loss": 0.81292975,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.83585477,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 2.6128122806549072
    },
    {
      "auxiliary_loss_clip": 0.01162139,
      "auxiliary_loss_mlp": 0.01021921,
      "balance_loss_clip": 0.9317013,
      "balance_loss_mlp": 1.01581144,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 1.8778178764530584,
      "language_loss": 0.73263568,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.75447631,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.6360416412353516
    },
    {
      "auxiliary_loss_clip": 0.01162372,
      "auxiliary_loss_mlp": 0.01029271,
      "balance_loss_clip": 1.04449081,
      "balance_loss_mlp": 1.02218425,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.7894220985774634,
      "language_loss": 0.84446537,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86638176,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.5955891609191895
    },
    {
      "auxiliary_loss_clip": 0.01166204,
      "auxiliary_loss_mlp": 0.01032596,
      "balance_loss_clip": 1.00994706,
      "balance_loss_mlp": 1.02582514,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 1.532344948963828,
      "language_loss": 0.66182649,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.68381453,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.5973496437072754
    },
    {
      "auxiliary_loss_clip": 0.01165819,
      "auxiliary_loss_mlp": 0.01023809,
      "balance_loss_clip": 0.97074008,
      "balance_loss_mlp": 1.01709735,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 1.6773859765418409,
      "language_loss": 0.77633744,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79823375,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.8281190395355225
    },
    {
      "auxiliary_loss_clip": 0.01164526,
      "auxiliary_loss_mlp": 0.0102186,
      "balance_loss_clip": 0.93580663,
      "balance_loss_mlp": 1.01501727,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 1.6331186117793872,
      "language_loss": 0.75938803,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78125191,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 3.551511764526367
    },
    {
      "auxiliary_loss_clip": 0.01164526,
      "auxiliary_loss_mlp": 0.01023914,
      "balance_loss_clip": 1.01170599,
      "balance_loss_mlp": 1.01682091,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 1.5460844711898005,
      "language_loss": 0.70260787,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.72449231,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.6536812782287598
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.01030596,
      "balance_loss_clip": 0.97273546,
      "balance_loss_mlp": 1.02355051,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 1.9047772344937972,
      "language_loss": 0.82390636,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84590375,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.682917594909668
    },
    {
      "auxiliary_loss_clip": 0.0117302,
      "auxiliary_loss_mlp": 0.01026893,
      "balance_loss_clip": 0.97093719,
      "balance_loss_mlp": 1.01947832,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 1.7302774670092613,
      "language_loss": 0.70017052,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.72216964,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 2.9201080799102783
    },
    {
      "auxiliary_loss_clip": 0.01165546,
      "auxiliary_loss_mlp": 0.01023479,
      "balance_loss_clip": 1.01093733,
      "balance_loss_mlp": 1.0165168,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 1.7865899820090545,
      "language_loss": 0.74264491,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76453519,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 2.6201670169830322
    },
    {
      "auxiliary_loss_clip": 0.01153502,
      "auxiliary_loss_mlp": 0.01122368,
      "balance_loss_clip": 0.96864426,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 1.9684303630202433,
      "language_loss": 0.64198202,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.66474068,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 3.6135640144348145
    },
    {
      "auxiliary_loss_clip": 0.01144415,
      "auxiliary_loss_mlp": 0.01019086,
      "balance_loss_clip": 0.9287163,
      "balance_loss_mlp": 1.01289606,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.5186185415682445,
      "language_loss": 0.74158061,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76321566,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.9316577911376953
    },
    {
      "auxiliary_loss_clip": 0.01158667,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 0.96585727,
      "balance_loss_mlp": 1.01964533,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 1.587130863687591,
      "language_loss": 0.80374932,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.82560337,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.720879554748535
    },
    {
      "auxiliary_loss_clip": 0.01155756,
      "auxiliary_loss_mlp": 0.01021659,
      "balance_loss_clip": 0.96744478,
      "balance_loss_mlp": 1.01477718,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 1.9002740991069982,
      "language_loss": 0.61035538,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.63212949,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 3.7881367206573486
    },
    {
      "auxiliary_loss_clip": 0.0115491,
      "auxiliary_loss_mlp": 0.01022811,
      "balance_loss_clip": 0.96593034,
      "balance_loss_mlp": 1.01589406,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.6094896024782854,
      "language_loss": 0.84358859,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86536574,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.7006008625030518
    },
    {
      "auxiliary_loss_clip": 0.01163317,
      "auxiliary_loss_mlp": 0.01122042,
      "balance_loss_clip": 1.00850403,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 1.8240489485898232,
      "language_loss": 0.64829218,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.6711458,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.6849162578582764
    },
    {
      "auxiliary_loss_clip": 0.01160565,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 0.93106943,
      "balance_loss_mlp": 1.01786304,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.383340822685766,
      "language_loss": 0.68849814,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.7103563,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 2.7326676845550537
    },
    {
      "auxiliary_loss_clip": 0.0116533,
      "auxiliary_loss_mlp": 0.01022031,
      "balance_loss_clip": 1.04783881,
      "balance_loss_mlp": 1.01528704,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 1.537332665940809,
      "language_loss": 0.70602727,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72790086,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.574253559112549
    },
    {
      "auxiliary_loss_clip": 0.0116271,
      "auxiliary_loss_mlp": 0.01033021,
      "balance_loss_clip": 1.00975132,
      "balance_loss_mlp": 1.02599978,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 1.833020576755294,
      "language_loss": 0.72259599,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74455333,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.6439807415008545
    },
    {
      "auxiliary_loss_clip": 0.01160466,
      "auxiliary_loss_mlp": 0.01122401,
      "balance_loss_clip": 1.00650048,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 1.837462177627161,
      "language_loss": 0.86877763,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.89160633,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 2.6147074699401855
    },
    {
      "auxiliary_loss_clip": 0.01150008,
      "auxiliary_loss_mlp": 0.01025788,
      "balance_loss_clip": 0.84832937,
      "balance_loss_mlp": 1.01884985,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 2.2674974063116373,
      "language_loss": 0.74526429,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.76702225,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 2.73209810256958
    },
    {
      "auxiliary_loss_clip": 0.01166139,
      "auxiliary_loss_mlp": 0.01021926,
      "balance_loss_clip": 0.97180849,
      "balance_loss_mlp": 1.01478565,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 1.820449017154609,
      "language_loss": 0.7426002,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76448083,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.6411514282226562
    },
    {
      "auxiliary_loss_clip": 0.01059948,
      "auxiliary_loss_mlp": 0.01001469,
      "balance_loss_clip": 0.97191471,
      "balance_loss_mlp": 0.99984783,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7711765361911551,
      "language_loss": 0.56365705,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58427119,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.1608800888061523
    },
    {
      "auxiliary_loss_clip": 0.01167774,
      "auxiliary_loss_mlp": 0.0102949,
      "balance_loss_clip": 1.01170397,
      "balance_loss_mlp": 1.0224719,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 1.5764859833176026,
      "language_loss": 0.78006017,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80203283,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 2.590207576751709
    },
    {
      "auxiliary_loss_clip": 0.01165623,
      "auxiliary_loss_mlp": 0.01024257,
      "balance_loss_clip": 1.04627848,
      "balance_loss_mlp": 1.01695836,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.4203600309997735,
      "language_loss": 0.67843401,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.70033288,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.7471678256988525
    },
    {
      "auxiliary_loss_clip": 0.01165831,
      "auxiliary_loss_mlp": 0.0101892,
      "balance_loss_clip": 1.01082969,
      "balance_loss_mlp": 1.01188326,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 1.8195741296823815,
      "language_loss": 0.76365221,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.78549969,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 2.5941617488861084
    },
    {
      "auxiliary_loss_clip": 0.01154418,
      "auxiliary_loss_mlp": 0.01027619,
      "balance_loss_clip": 0.96683729,
      "balance_loss_mlp": 1.02032363,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 1.533098452529569,
      "language_loss": 0.70473641,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72655672,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.7897074222564697
    },
    {
      "auxiliary_loss_clip": 0.01170383,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.04911804,
      "balance_loss_mlp": 1.01909375,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.28643715037375,
      "language_loss": 0.82261586,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84458083,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 2.576186180114746
    },
    {
      "auxiliary_loss_clip": 0.01159128,
      "auxiliary_loss_mlp": 0.01023204,
      "balance_loss_clip": 0.97140008,
      "balance_loss_mlp": 1.01623046,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 2.502962751282139,
      "language_loss": 0.74693799,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.7687614,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.6198651790618896
    },
    {
      "auxiliary_loss_clip": 0.01152253,
      "auxiliary_loss_mlp": 0.01121692,
      "balance_loss_clip": 0.85262835,
      "balance_loss_mlp": 0.0,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 1.675587438055266,
      "language_loss": 0.78175718,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.80449665,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.7384541034698486
    },
    {
      "auxiliary_loss_clip": 0.01161528,
      "auxiliary_loss_mlp": 0.01031457,
      "balance_loss_clip": 0.89462471,
      "balance_loss_mlp": 1.02420354,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.5846917393297395,
      "language_loss": 0.82990795,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85183775,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.68365216255188
    },
    {
      "auxiliary_loss_clip": 0.01166239,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 1.04699028,
      "balance_loss_mlp": 1.01931262,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 1.8404798022363273,
      "language_loss": 0.77540278,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.79732347,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 3.3656539916992188
    },
    {
      "auxiliary_loss_clip": 0.01168817,
      "auxiliary_loss_mlp": 0.01023332,
      "balance_loss_clip": 1.00994503,
      "balance_loss_mlp": 1.01635504,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.1155150638384557,
      "language_loss": 0.80401397,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82593548,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.612881898880005
    },
    {
      "auxiliary_loss_clip": 0.01173459,
      "auxiliary_loss_mlp": 0.01028935,
      "balance_loss_clip": 0.9762947,
      "balance_loss_mlp": 1.02131748,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 1.8531618558737075,
      "language_loss": 0.76774353,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.7897675,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.6234967708587646
    },
    {
      "auxiliary_loss_clip": 0.01171247,
      "auxiliary_loss_mlp": 0.01031889,
      "balance_loss_clip": 0.9720701,
      "balance_loss_mlp": 1.02470064,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 3.150229896697412,
      "language_loss": 0.7564981,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.77852947,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.6234281063079834
    },
    {
      "auxiliary_loss_clip": 0.01166532,
      "auxiliary_loss_mlp": 0.01122339,
      "balance_loss_clip": 1.04691911,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.6082601357793935,
      "language_loss": 0.80842805,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.83131683,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 3.6017541885375977
    },
    {
      "auxiliary_loss_clip": 0.01056751,
      "auxiliary_loss_mlp": 0.01002868,
      "balance_loss_clip": 0.97023773,
      "balance_loss_mlp": 1.00122249,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8704535792367987,
      "language_loss": 0.6098938,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.63048995,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.1566081047058105
    },
    {
      "auxiliary_loss_clip": 0.0115453,
      "auxiliary_loss_mlp": 0.01028315,
      "balance_loss_clip": 0.92694783,
      "balance_loss_mlp": 1.02076292,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 1.905574890825056,
      "language_loss": 0.78900814,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.81083655,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.689258337020874
    },
    {
      "auxiliary_loss_clip": 0.01164206,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 0.92969042,
      "balance_loss_mlp": 1.01778913,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.2730254025978684,
      "language_loss": 0.69013977,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.7120294,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.729212999343872
    },
    {
      "auxiliary_loss_clip": 0.01166201,
      "auxiliary_loss_mlp": 0.01122799,
      "balance_loss_clip": 1.04808331,
      "balance_loss_mlp": 0.0,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 1.7382100271190541,
      "language_loss": 0.88269973,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.9055897,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 3.4791629314422607
    },
    {
      "auxiliary_loss_clip": 0.01168813,
      "auxiliary_loss_mlp": 0.01122156,
      "balance_loss_clip": 1.01071715,
      "balance_loss_mlp": 0.0,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 1.7668532520030544,
      "language_loss": 0.74474764,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76765734,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 3.0835001468658447
    },
    {
      "auxiliary_loss_clip": 0.011622,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 0.93397653,
      "balance_loss_mlp": 1.02065623,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 2.3829045924425403,
      "language_loss": 0.72370422,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.74560928,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.66111421585083
    },
    {
      "auxiliary_loss_clip": 0.01166932,
      "auxiliary_loss_mlp": 0.01021605,
      "balance_loss_clip": 0.93229347,
      "balance_loss_mlp": 1.01446176,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 1.9880348293219967,
      "language_loss": 0.72362578,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74551117,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 2.7036378383636475
    },
    {
      "auxiliary_loss_clip": 0.01164008,
      "auxiliary_loss_mlp": 0.01021226,
      "balance_loss_clip": 1.04689467,
      "balance_loss_mlp": 1.0144074,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 1.5212976495724722,
      "language_loss": 0.6388526,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.66070497,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 2.6182334423065186
    },
    {
      "auxiliary_loss_clip": 0.01160658,
      "auxiliary_loss_mlp": 0.01024939,
      "balance_loss_clip": 0.92871261,
      "balance_loss_mlp": 1.01832032,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 2.3343550218128697,
      "language_loss": 0.70032895,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.7221849,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.66542387008667
    },
    {
      "auxiliary_loss_clip": 0.01170248,
      "auxiliary_loss_mlp": 0.01028328,
      "balance_loss_clip": 1.01295972,
      "balance_loss_mlp": 1.02030563,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 1.9559929407807846,
      "language_loss": 0.79533154,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.81731731,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 2.588927745819092
    },
    {
      "auxiliary_loss_clip": 0.01169389,
      "auxiliary_loss_mlp": 0.01122081,
      "balance_loss_clip": 1.00988007,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.354143824657775,
      "language_loss": 0.64166528,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.66457999,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.7166168689727783
    },
    {
      "auxiliary_loss_clip": 0.01164979,
      "auxiliary_loss_mlp": 0.01024526,
      "balance_loss_clip": 0.97189677,
      "balance_loss_mlp": 1.01697373,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 2.414713357104854,
      "language_loss": 0.71101749,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73291254,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.6721155643463135
    },
    {
      "auxiliary_loss_clip": 0.01167642,
      "auxiliary_loss_mlp": 0.01025494,
      "balance_loss_clip": 0.93254811,
      "balance_loss_mlp": 1.01830244,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 1.8756154784890728,
      "language_loss": 0.62241143,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64434278,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.6381940841674805
    },
    {
      "auxiliary_loss_clip": 0.01173558,
      "auxiliary_loss_mlp": 0.01025699,
      "balance_loss_clip": 0.97108305,
      "balance_loss_mlp": 1.0180012,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 3.776643417486623,
      "language_loss": 0.70073164,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.7227242,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 2.687178611755371
    },
    {
      "auxiliary_loss_clip": 0.01165687,
      "auxiliary_loss_mlp": 0.01026565,
      "balance_loss_clip": 0.97030109,
      "balance_loss_mlp": 1.01904845,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 2.477705238212912,
      "language_loss": 0.74751091,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.76943338,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.6079068183898926
    },
    {
      "auxiliary_loss_clip": 0.01062585,
      "auxiliary_loss_mlp": 0.01003948,
      "balance_loss_clip": 0.97042531,
      "balance_loss_mlp": 1.00237465,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7172644939825148,
      "language_loss": 0.56321871,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.583884,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.2450313568115234
    },
    {
      "auxiliary_loss_clip": 0.01163787,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.04500508,
      "balance_loss_mlp": 1.01822114,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 1.6071264729608863,
      "language_loss": 0.67727768,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.69916797,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 2.5913898944854736
    },
    {
      "auxiliary_loss_clip": 0.01166712,
      "auxiliary_loss_mlp": 0.01020133,
      "balance_loss_clip": 1.04773128,
      "balance_loss_mlp": 1.01328754,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 1.9144959631518657,
      "language_loss": 0.79845858,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.82032704,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.540355920791626
    },
    {
      "auxiliary_loss_clip": 0.01165948,
      "auxiliary_loss_mlp": 0.01022903,
      "balance_loss_clip": 1.0486505,
      "balance_loss_mlp": 1.01614356,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 2.421072442230197,
      "language_loss": 0.69939983,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72128832,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 2.6127421855926514
    },
    {
      "auxiliary_loss_clip": 0.0116204,
      "auxiliary_loss_mlp": 0.01029483,
      "balance_loss_clip": 0.96902394,
      "balance_loss_mlp": 1.02215719,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 1.8782233638193424,
      "language_loss": 0.71469116,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.73660648,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.5750253200531006
    },
    {
      "auxiliary_loss_clip": 0.01168038,
      "auxiliary_loss_mlp": 0.01024763,
      "balance_loss_clip": 0.97303045,
      "balance_loss_mlp": 1.01724708,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 3.475011364859161,
      "language_loss": 0.79367554,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81560349,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 2.622039556503296
    },
    {
      "auxiliary_loss_clip": 0.01163698,
      "auxiliary_loss_mlp": 0.01025771,
      "balance_loss_clip": 0.97273946,
      "balance_loss_mlp": 1.01856184,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 4.027660452961855,
      "language_loss": 0.75807959,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.77997434,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 3.4913363456726074
    },
    {
      "auxiliary_loss_clip": 0.01164962,
      "auxiliary_loss_mlp": 0.01026437,
      "balance_loss_clip": 1.01036358,
      "balance_loss_mlp": 1.0195291,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 1.9724445444361414,
      "language_loss": 0.69468141,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.71659541,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.7514469623565674
    },
    {
      "auxiliary_loss_clip": 0.01161243,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 0.97045231,
      "balance_loss_mlp": 1.02039266,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.690605112772439,
      "language_loss": 0.66423243,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.6861226,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.687032699584961
    },
    {
      "auxiliary_loss_clip": 0.01163142,
      "auxiliary_loss_mlp": 0.01026368,
      "balance_loss_clip": 0.97003543,
      "balance_loss_mlp": 1.01985574,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 2.1454465297165117,
      "language_loss": 0.80463696,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82653207,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.6113786697387695
    },
    {
      "auxiliary_loss_clip": 0.01165607,
      "auxiliary_loss_mlp": 0.01027298,
      "balance_loss_clip": 1.00759816,
      "balance_loss_mlp": 1.01965356,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 2.8653109231524154,
      "language_loss": 0.73009503,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75202411,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 3.5320358276367188
    },
    {
      "auxiliary_loss_clip": 0.01167805,
      "auxiliary_loss_mlp": 0.01026498,
      "balance_loss_clip": 0.97133255,
      "balance_loss_mlp": 1.02013516,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 1.4995711095864195,
      "language_loss": 0.72463471,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74657774,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 2.6795120239257812
    },
    {
      "auxiliary_loss_clip": 0.01163992,
      "auxiliary_loss_mlp": 0.01024103,
      "balance_loss_clip": 0.93320286,
      "balance_loss_mlp": 1.01745689,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 1.8832922703018216,
      "language_loss": 0.7054683,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.72734928,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.6641812324523926
    },
    {
      "auxiliary_loss_clip": 0.01166253,
      "auxiliary_loss_mlp": 0.01122575,
      "balance_loss_clip": 1.00967932,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 1.677715940360335,
      "language_loss": 0.83805716,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.86094546,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.57597279548645
    },
    {
      "auxiliary_loss_clip": 0.01069313,
      "auxiliary_loss_mlp": 0.01003949,
      "balance_loss_clip": 0.89920568,
      "balance_loss_mlp": 1.00237513,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.8437211050439556,
      "language_loss": 0.56013441,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.58086705,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 4.931986331939697
    },
    {
      "auxiliary_loss_clip": 0.01168205,
      "auxiliary_loss_mlp": 0.01028156,
      "balance_loss_clip": 0.85530198,
      "balance_loss_mlp": 1.02090478,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 2.320494503022459,
      "language_loss": 0.80625165,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.82821524,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.818499803543091
    },
    {
      "auxiliary_loss_clip": 0.01152002,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 0.92941713,
      "balance_loss_mlp": 1.02158868,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 1.9724912482431285,
      "language_loss": 0.81600434,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.83781183,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.830515146255493
    },
    {
      "auxiliary_loss_clip": 0.01161785,
      "auxiliary_loss_mlp": 0.01019436,
      "balance_loss_clip": 0.96899474,
      "balance_loss_mlp": 1.01284397,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 2.4406781112564837,
      "language_loss": 0.69636482,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.71817708,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 2.655310869216919
    },
    {
      "auxiliary_loss_clip": 0.01169873,
      "auxiliary_loss_mlp": 0.01027577,
      "balance_loss_clip": 0.97370255,
      "balance_loss_mlp": 1.02048099,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 4.964867023106709,
      "language_loss": 0.64244032,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66441482,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 2.8160130977630615
    },
    {
      "auxiliary_loss_clip": 0.01163987,
      "auxiliary_loss_mlp": 0.01122193,
      "balance_loss_clip": 0.97063911,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.6015821100485372,
      "language_loss": 0.80488449,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.82774627,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.7528135776519775
    },
    {
      "auxiliary_loss_clip": 0.0115441,
      "auxiliary_loss_mlp": 0.01024977,
      "balance_loss_clip": 0.97008359,
      "balance_loss_mlp": 1.01846862,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 1.7663305407532732,
      "language_loss": 0.73828506,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.76007891,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 2.6497576236724854
    },
    {
      "auxiliary_loss_clip": 0.01058321,
      "auxiliary_loss_mlp": 0.01003196,
      "balance_loss_clip": 0.93533766,
      "balance_loss_mlp": 1.00163424,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8531718722501772,
      "language_loss": 0.62256891,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64318407,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.1681466102600098
    },
    {
      "auxiliary_loss_clip": 0.01165357,
      "auxiliary_loss_mlp": 0.01026935,
      "balance_loss_clip": 1.04694164,
      "balance_loss_mlp": 1.02027416,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.0066434699071527,
      "language_loss": 0.75123888,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77316177,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.616074323654175
    },
    {
      "auxiliary_loss_clip": 0.01166979,
      "auxiliary_loss_mlp": 0.01024355,
      "balance_loss_clip": 1.00957692,
      "balance_loss_mlp": 1.01738095,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 2.347514205903049,
      "language_loss": 0.83384812,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85576141,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 2.578315258026123
    },
    {
      "auxiliary_loss_clip": 0.01164613,
      "auxiliary_loss_mlp": 0.01025614,
      "balance_loss_clip": 0.97413242,
      "balance_loss_mlp": 1.01828599,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 2.8560791713653666,
      "language_loss": 0.84547335,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.86737561,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.607670783996582
    },
    {
      "auxiliary_loss_clip": 0.01164676,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.00879049,
      "balance_loss_mlp": 1.01969266,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 6.544812322861209,
      "language_loss": 0.86043215,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88235033,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.5868642330169678
    },
    {
      "auxiliary_loss_clip": 0.01150275,
      "auxiliary_loss_mlp": 0.0102498,
      "balance_loss_clip": 0.85314608,
      "balance_loss_mlp": 1.01788378,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 1.7776824030694782,
      "language_loss": 0.75738227,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.77913475,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.838446617126465
    },
    {
      "auxiliary_loss_clip": 0.01160605,
      "auxiliary_loss_mlp": 0.01028613,
      "balance_loss_clip": 0.89599818,
      "balance_loss_mlp": 1.02199411,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.5488431573391506,
      "language_loss": 0.76397038,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78586257,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 2.655745506286621
    },
    {
      "auxiliary_loss_clip": 0.01171419,
      "auxiliary_loss_mlp": 0.01025152,
      "balance_loss_clip": 0.96918535,
      "balance_loss_mlp": 1.01786804,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 1.8568156827376798,
      "language_loss": 0.75804764,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.78001338,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.7016263008117676
    },
    {
      "auxiliary_loss_clip": 0.01166104,
      "auxiliary_loss_mlp": 0.01022172,
      "balance_loss_clip": 1.05064607,
      "balance_loss_mlp": 1.01573813,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 1.8581569717462314,
      "language_loss": 0.73492789,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75681067,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 2.794764518737793
    },
    {
      "auxiliary_loss_clip": 0.01164998,
      "auxiliary_loss_mlp": 0.01025585,
      "balance_loss_clip": 0.96850133,
      "balance_loss_mlp": 1.01804745,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 2.397287453617307,
      "language_loss": 0.67546797,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69737375,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.6717989444732666
    },
    {
      "auxiliary_loss_clip": 0.0116328,
      "auxiliary_loss_mlp": 0.0102758,
      "balance_loss_clip": 1.0088563,
      "balance_loss_mlp": 1.02039123,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.1632497953724408,
      "language_loss": 0.84972322,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87163174,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 2.6001100540161133
    },
    {
      "auxiliary_loss_clip": 0.01170957,
      "auxiliary_loss_mlp": 0.01023123,
      "balance_loss_clip": 0.85519326,
      "balance_loss_mlp": 1.01604545,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 1.9028634722365994,
      "language_loss": 0.73623818,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.75817895,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 3.545804738998413
    },
    {
      "auxiliary_loss_clip": 0.01162646,
      "auxiliary_loss_mlp": 0.01122165,
      "balance_loss_clip": 0.97091937,
      "balance_loss_mlp": 0.0,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 1.6735335304419956,
      "language_loss": 0.71325576,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.73610389,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.6042284965515137
    },
    {
      "auxiliary_loss_clip": 0.01164094,
      "auxiliary_loss_mlp": 0.01025862,
      "balance_loss_clip": 1.00901747,
      "balance_loss_mlp": 1.01863766,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 1.8258816362610895,
      "language_loss": 0.82556343,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.84746301,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 2.6397194862365723
    },
    {
      "auxiliary_loss_clip": 0.01165071,
      "auxiliary_loss_mlp": 0.01025365,
      "balance_loss_clip": 1.00868368,
      "balance_loss_mlp": 1.01842141,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 1.6218821129830132,
      "language_loss": 0.80763173,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.82953608,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.627126932144165
    },
    {
      "auxiliary_loss_clip": 0.01159441,
      "auxiliary_loss_mlp": 0.01025851,
      "balance_loss_clip": 1.00799906,
      "balance_loss_mlp": 1.0185132,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 1.6607054772424086,
      "language_loss": 0.76955438,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79140735,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.551815986633301
    },
    {
      "auxiliary_loss_clip": 0.01155764,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 0.96960688,
      "balance_loss_mlp": 1.01917922,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 2.49760370608401,
      "language_loss": 0.74573445,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76755381,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 3.6033735275268555
    },
    {
      "auxiliary_loss_clip": 0.01152445,
      "auxiliary_loss_mlp": 0.01026933,
      "balance_loss_clip": 0.8532787,
      "balance_loss_mlp": 1.01998949,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 2.1070698850598824,
      "language_loss": 0.72604692,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74784076,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 2.7496461868286133
    },
    {
      "auxiliary_loss_clip": 0.01156757,
      "auxiliary_loss_mlp": 0.01027777,
      "balance_loss_clip": 0.96860451,
      "balance_loss_mlp": 1.0203743,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 1.8197176152980863,
      "language_loss": 0.74905169,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77089703,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 2.6783876419067383
    },
    {
      "auxiliary_loss_clip": 0.01160598,
      "auxiliary_loss_mlp": 0.01026118,
      "balance_loss_clip": 0.89159876,
      "balance_loss_mlp": 1.01933765,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 2.628937656088704,
      "language_loss": 0.78322965,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80509686,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 3.6893744468688965
    },
    {
      "auxiliary_loss_clip": 0.01169138,
      "auxiliary_loss_mlp": 0.01122995,
      "balance_loss_clip": 1.04590034,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 1.941957440504079,
      "language_loss": 0.79777247,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.82069379,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 3.4692001342773438
    },
    {
      "auxiliary_loss_clip": 0.01167459,
      "auxiliary_loss_mlp": 0.0102363,
      "balance_loss_clip": 0.9733367,
      "balance_loss_mlp": 1.01643014,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 2.900892926469116,
      "language_loss": 0.70742738,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.72933823,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.7873783111572266
    },
    {
      "auxiliary_loss_clip": 0.01155828,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 0.92915642,
      "balance_loss_mlp": 1.02087915,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.2064867318233823,
      "language_loss": 0.77046353,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.7923007,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.6296138763427734
    },
    {
      "auxiliary_loss_clip": 0.0116803,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 1.04893422,
      "balance_loss_mlp": 1.01713037,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 1.5808612342739634,
      "language_loss": 0.73967868,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.76159561,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 2.6200127601623535
    },
    {
      "auxiliary_loss_clip": 0.01058969,
      "auxiliary_loss_mlp": 0.0100202,
      "balance_loss_clip": 1.00811362,
      "balance_loss_mlp": 1.00039887,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.9537793900978883,
      "language_loss": 0.64549017,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66610008,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.2081480026245117
    },
    {
      "auxiliary_loss_clip": 0.01164308,
      "auxiliary_loss_mlp": 0.01028463,
      "balance_loss_clip": 0.97265249,
      "balance_loss_mlp": 1.02118254,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.0199473763494096,
      "language_loss": 0.76797277,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.78990048,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.6157612800598145
    },
    {
      "auxiliary_loss_clip": 0.01167703,
      "auxiliary_loss_mlp": 0.01028711,
      "balance_loss_clip": 0.89519918,
      "balance_loss_mlp": 1.02118266,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.4749748646673149,
      "language_loss": 0.83780682,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.85977101,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 2.6919400691986084
    },
    {
      "auxiliary_loss_clip": 0.01170476,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 1.04869521,
      "balance_loss_mlp": 1.01747131,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 2.793412101821808,
      "language_loss": 0.6661731,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.6881299,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.6910345554351807
    },
    {
      "auxiliary_loss_clip": 0.01159295,
      "auxiliary_loss_mlp": 0.01027767,
      "balance_loss_clip": 0.96602982,
      "balance_loss_mlp": 1.02039969,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 2.744543137124113,
      "language_loss": 0.68449485,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.70636547,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 2.6285183429718018
    },
    {
      "auxiliary_loss_clip": 0.01161908,
      "auxiliary_loss_mlp": 0.01021853,
      "balance_loss_clip": 0.92958403,
      "balance_loss_mlp": 1.01449823,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 2.010682438008242,
      "language_loss": 0.73549753,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.75733519,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.668895959854126
    },
    {
      "auxiliary_loss_clip": 0.0116652,
      "auxiliary_loss_mlp": 0.01023295,
      "balance_loss_clip": 0.96976632,
      "balance_loss_mlp": 1.01638067,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 2.050640548561514,
      "language_loss": 0.71806121,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.73995936,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 2.641554117202759
    },
    {
      "auxiliary_loss_clip": 0.01164297,
      "auxiliary_loss_mlp": 0.01022374,
      "balance_loss_clip": 1.00754952,
      "balance_loss_mlp": 1.01588631,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 2.2111491782846726,
      "language_loss": 0.78408372,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.8059504,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.63655161857605
    },
    {
      "auxiliary_loss_clip": 0.01159227,
      "auxiliary_loss_mlp": 0.01027092,
      "balance_loss_clip": 0.97035396,
      "balance_loss_mlp": 1.02018654,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 3.749491845322591,
      "language_loss": 0.69103098,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71289414,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.727937698364258
    },
    {
      "auxiliary_loss_clip": 0.01162205,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 0.96909761,
      "balance_loss_mlp": 1.02114081,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 1.80114311967172,
      "language_loss": 0.75187773,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.77378005,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.631484270095825
    },
    {
      "auxiliary_loss_clip": 0.01155922,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 0.93373752,
      "balance_loss_mlp": 1.02057862,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 1.8730945595961792,
      "language_loss": 0.78558958,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.80742192,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.6950933933258057
    },
    {
      "auxiliary_loss_clip": 0.0115441,
      "auxiliary_loss_mlp": 0.01024835,
      "balance_loss_clip": 0.96690404,
      "balance_loss_mlp": 1.01814461,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 2.5737610316502795,
      "language_loss": 0.78472471,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.80651718,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 2.596085786819458
    },
    {
      "auxiliary_loss_clip": 0.01166478,
      "auxiliary_loss_mlp": 0.01122364,
      "balance_loss_clip": 1.00679243,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 1.809032858387605,
      "language_loss": 0.73382163,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.75671005,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.6682913303375244
    },
    {
      "auxiliary_loss_clip": 0.0116466,
      "auxiliary_loss_mlp": 0.01031446,
      "balance_loss_clip": 0.93916947,
      "balance_loss_mlp": 1.02398682,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 1.9564825007942768,
      "language_loss": 0.80703962,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.82900071,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 3.5806994438171387
    },
    {
      "auxiliary_loss_clip": 0.01156399,
      "auxiliary_loss_mlp": 0.01021276,
      "balance_loss_clip": 0.89170444,
      "balance_loss_mlp": 1.01454329,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 2.8822770250370517,
      "language_loss": 0.81433499,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.83611172,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.7220616340637207
    },
    {
      "auxiliary_loss_clip": 0.01157893,
      "auxiliary_loss_mlp": 0.01024003,
      "balance_loss_clip": 0.93026757,
      "balance_loss_mlp": 1.01680303,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 1.4123433666810181,
      "language_loss": 0.77277684,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79459584,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.9162447452545166
    },
    {
      "auxiliary_loss_clip": 0.01067379,
      "auxiliary_loss_mlp": 0.01001827,
      "balance_loss_clip": 0.93511641,
      "balance_loss_mlp": 1.0002532,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.7429162324851339,
      "language_loss": 0.53607708,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.55676919,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 3.284099578857422
    },
    {
      "auxiliary_loss_clip": 0.01163565,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.01059294,
      "balance_loss_mlp": 1.02013361,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 1.7156054626714785,
      "language_loss": 0.7523495,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77425313,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 3.4691905975341797
    },
    {
      "auxiliary_loss_clip": 0.01158321,
      "auxiliary_loss_mlp": 0.01027527,
      "balance_loss_clip": 1.00757062,
      "balance_loss_mlp": 1.01998365,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 1.7058332348926641,
      "language_loss": 0.76507366,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.78693211,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.7087225914001465
    },
    {
      "auxiliary_loss_clip": 0.01163857,
      "auxiliary_loss_mlp": 0.01030006,
      "balance_loss_clip": 1.01162672,
      "balance_loss_mlp": 1.02307987,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 3.198249636475753,
      "language_loss": 0.68817174,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.71011043,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 2.778054714202881
    },
    {
      "auxiliary_loss_clip": 0.011647,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 0.89505804,
      "balance_loss_mlp": 1.02214313,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 1.86648778687186,
      "language_loss": 0.64384842,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66578615,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 2.7849748134613037
    },
    {
      "auxiliary_loss_clip": 0.01164288,
      "auxiliary_loss_mlp": 0.0102556,
      "balance_loss_clip": 1.04781294,
      "balance_loss_mlp": 1.01918805,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.817888274262644,
      "language_loss": 0.80765992,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.82955843,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 3.4806723594665527
    },
    {
      "auxiliary_loss_clip": 0.01168375,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.05015028,
      "balance_loss_mlp": 1.0183835,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 1.9562628212029012,
      "language_loss": 0.78525966,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.80719692,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.576110363006592
    },
    {
      "auxiliary_loss_clip": 0.0114953,
      "auxiliary_loss_mlp": 0.01023102,
      "balance_loss_clip": 0.93070763,
      "balance_loss_mlp": 1.01667356,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.982877076897686,
      "language_loss": 0.73534149,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.7570678,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.6824564933776855
    },
    {
      "auxiliary_loss_clip": 0.01151641,
      "auxiliary_loss_mlp": 0.01122542,
      "balance_loss_clip": 0.96710408,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 1.7469506262738672,
      "language_loss": 0.70408499,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.72682679,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.6920571327209473
    },
    {
      "auxiliary_loss_clip": 0.01170752,
      "auxiliary_loss_mlp": 0.01024733,
      "balance_loss_clip": 0.9332149,
      "balance_loss_mlp": 1.01735377,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 1.9377385721917508,
      "language_loss": 0.66080785,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68276274,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.6650655269622803
    },
    {
      "auxiliary_loss_clip": 0.01164008,
      "auxiliary_loss_mlp": 0.01027467,
      "balance_loss_clip": 1.01074922,
      "balance_loss_mlp": 1.02063584,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 2.200800944213822,
      "language_loss": 0.74880999,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.77072471,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.561868906021118
    },
    {
      "auxiliary_loss_clip": 0.01160068,
      "auxiliary_loss_mlp": 0.01024044,
      "balance_loss_clip": 0.9704994,
      "balance_loss_mlp": 1.01726389,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.6123880946828224,
      "language_loss": 0.71315956,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73500061,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.7231192588806152
    },
    {
      "auxiliary_loss_clip": 0.01067531,
      "auxiliary_loss_mlp": 0.01000782,
      "balance_loss_clip": 0.9351747,
      "balance_loss_mlp": 0.99914879,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8318979335580811,
      "language_loss": 0.53437197,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55505514,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 3.282475709915161
    },
    {
      "auxiliary_loss_clip": 0.01156253,
      "auxiliary_loss_mlp": 0.01026027,
      "balance_loss_clip": 0.93570876,
      "balance_loss_mlp": 1.01918721,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.5388334472871716,
      "language_loss": 0.77592081,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.79774356,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.7477834224700928
    },
    {
      "auxiliary_loss_clip": 0.01153681,
      "auxiliary_loss_mlp": 0.01023635,
      "balance_loss_clip": 0.96499807,
      "balance_loss_mlp": 1.01705456,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 1.887407710668317,
      "language_loss": 0.72109413,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.74286723,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 2.7090113162994385
    },
    {
      "auxiliary_loss_clip": 0.01170467,
      "auxiliary_loss_mlp": 0.01023884,
      "balance_loss_clip": 0.93433511,
      "balance_loss_mlp": 1.01630187,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.5355701416598952,
      "language_loss": 0.69570905,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.71765256,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.654324531555176
    },
    {
      "auxiliary_loss_clip": 0.0116786,
      "auxiliary_loss_mlp": 0.01027254,
      "balance_loss_clip": 0.97373486,
      "balance_loss_mlp": 1.01987505,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 1.4425252552950623,
      "language_loss": 0.81589544,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83784652,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 2.6672723293304443
    },
    {
      "auxiliary_loss_clip": 0.01163871,
      "auxiliary_loss_mlp": 0.01025317,
      "balance_loss_clip": 0.97247446,
      "balance_loss_mlp": 1.0182271,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 2.0372135365668975,
      "language_loss": 0.85584223,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.87773418,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.6690449714660645
    },
    {
      "auxiliary_loss_clip": 0.01058901,
      "auxiliary_loss_mlp": 0.01000837,
      "balance_loss_clip": 1.00815344,
      "balance_loss_mlp": 0.99922746,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.9075183513866212,
      "language_loss": 0.61484528,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63544273,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.2070860862731934
    },
    {
      "auxiliary_loss_clip": 0.01062905,
      "auxiliary_loss_mlp": 0.01002228,
      "balance_loss_clip": 0.89856255,
      "balance_loss_mlp": 1.00067842,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7847211933498809,
      "language_loss": 0.58590555,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60655689,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.1366055011749268
    },
    {
      "auxiliary_loss_clip": 0.01166705,
      "auxiliary_loss_mlp": 0.01023745,
      "balance_loss_clip": 1.01034403,
      "balance_loss_mlp": 1.01639318,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.1586460276996036,
      "language_loss": 0.84813941,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.87004393,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.6361336708068848
    },
    {
      "auxiliary_loss_clip": 0.01158695,
      "auxiliary_loss_mlp": 0.01027929,
      "balance_loss_clip": 0.97012305,
      "balance_loss_mlp": 1.02106285,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 2.02937328704934,
      "language_loss": 0.7939868,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.815853,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 2.6236720085144043
    },
    {
      "auxiliary_loss_clip": 0.01163089,
      "auxiliary_loss_mlp": 0.01024192,
      "balance_loss_clip": 0.97040099,
      "balance_loss_mlp": 1.01755846,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.0025988755916377,
      "language_loss": 0.79695779,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.81883061,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 2.733466386795044
    },
    {
      "auxiliary_loss_clip": 0.01159319,
      "auxiliary_loss_mlp": 0.01026149,
      "balance_loss_clip": 0.93002486,
      "balance_loss_mlp": 1.01841807,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 1.8416728392397803,
      "language_loss": 0.72231156,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74416625,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.7725672721862793
    },
    {
      "auxiliary_loss_clip": 0.01161639,
      "auxiliary_loss_mlp": 0.01025621,
      "balance_loss_clip": 0.97069705,
      "balance_loss_mlp": 1.01817644,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.958353511818494,
      "language_loss": 0.67509711,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.69696975,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 4.105653285980225
    },
    {
      "auxiliary_loss_clip": 0.01164765,
      "auxiliary_loss_mlp": 0.01022074,
      "balance_loss_clip": 0.96782333,
      "balance_loss_mlp": 1.0153451,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.4837809498248475,
      "language_loss": 0.77164006,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79350853,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.7136073112487793
    },
    {
      "auxiliary_loss_clip": 0.01163978,
      "auxiliary_loss_mlp": 0.01020007,
      "balance_loss_clip": 1.00825906,
      "balance_loss_mlp": 1.01296496,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 1.7864722043288532,
      "language_loss": 0.86546314,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.88730299,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 2.6092448234558105
    },
    {
      "auxiliary_loss_clip": 0.01168761,
      "auxiliary_loss_mlp": 0.01022693,
      "balance_loss_clip": 1.0094018,
      "balance_loss_mlp": 1.01542759,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 1.8296134305341216,
      "language_loss": 0.78096604,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80288064,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 2.627180337905884
    },
    {
      "auxiliary_loss_clip": 0.01157014,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 0.96884364,
      "balance_loss_mlp": 1.02157235,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 1.9196836317779853,
      "language_loss": 0.77554131,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.79739726,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 3.6882517337799072
    },
    {
      "auxiliary_loss_clip": 0.01163542,
      "auxiliary_loss_mlp": 0.01025911,
      "balance_loss_clip": 0.96887523,
      "balance_loss_mlp": 1.01909804,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 2.61760838901501,
      "language_loss": 0.67560911,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.69750357,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.7434072494506836
    },
    {
      "auxiliary_loss_clip": 0.01164027,
      "auxiliary_loss_mlp": 0.01025523,
      "balance_loss_clip": 0.96863878,
      "balance_loss_mlp": 1.01849306,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.5223379029611621,
      "language_loss": 0.642564,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66445947,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 3.047553300857544
    },
    {
      "auxiliary_loss_clip": 0.01164762,
      "auxiliary_loss_mlp": 0.01024804,
      "balance_loss_clip": 0.93315971,
      "balance_loss_mlp": 1.01738286,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 1.8686238267137365,
      "language_loss": 0.71358281,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.73547846,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 4.483531951904297
    },
    {
      "auxiliary_loss_clip": 0.01166271,
      "auxiliary_loss_mlp": 0.01022921,
      "balance_loss_clip": 0.89281201,
      "balance_loss_mlp": 1.01620972,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 1.8762804703803644,
      "language_loss": 0.74128687,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.76317877,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.6735382080078125
    },
    {
      "auxiliary_loss_clip": 0.01167841,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 0.96870208,
      "balance_loss_mlp": 1.02486765,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 3.6133498919524163,
      "language_loss": 0.73531419,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.75731254,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.6517422199249268
    },
    {
      "auxiliary_loss_clip": 0.01063758,
      "auxiliary_loss_mlp": 0.01001385,
      "balance_loss_clip": 0.97146577,
      "balance_loss_mlp": 0.99971563,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.768696016656809,
      "language_loss": 0.53517008,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55582154,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.181544542312622
    },
    {
      "auxiliary_loss_clip": 0.01168256,
      "auxiliary_loss_mlp": 0.01123342,
      "balance_loss_clip": 1.04681849,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 2.1069216850800725,
      "language_loss": 0.79070288,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.81361884,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 2.804313898086548
    },
    {
      "auxiliary_loss_clip": 0.01166001,
      "auxiliary_loss_mlp": 0.01122302,
      "balance_loss_clip": 0.97374064,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 1.9168618381466622,
      "language_loss": 0.80347437,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.82635736,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.6650218963623047
    },
    {
      "auxiliary_loss_clip": 0.01163213,
      "auxiliary_loss_mlp": 0.01029878,
      "balance_loss_clip": 1.00747514,
      "balance_loss_mlp": 1.022663,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 1.572517887409946,
      "language_loss": 0.85560453,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87753546,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.634571075439453
    },
    {
      "auxiliary_loss_clip": 0.01160204,
      "auxiliary_loss_mlp": 0.01025059,
      "balance_loss_clip": 0.96798587,
      "balance_loss_mlp": 1.01833844,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 1.547490215766706,
      "language_loss": 0.65927672,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68112934,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 2.6802849769592285
    },
    {
      "auxiliary_loss_clip": 0.01144481,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 0.85004842,
      "balance_loss_mlp": 1.01729465,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 1.8089262424815102,
      "language_loss": 0.73153305,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.75323105,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.7514615058898926
    },
    {
      "auxiliary_loss_clip": 0.01173199,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.01145077,
      "balance_loss_mlp": 1.02228165,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.3224130762022184,
      "language_loss": 0.7304377,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75246036,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 2.7027711868286133
    },
    {
      "auxiliary_loss_clip": 0.01164077,
      "auxiliary_loss_mlp": 0.01021329,
      "balance_loss_clip": 0.9720962,
      "balance_loss_mlp": 1.01456344,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 2.2594668211192936,
      "language_loss": 0.71031725,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.73217124,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 2.6524407863616943
    },
    {
      "auxiliary_loss_clip": 0.01166219,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 0.97337568,
      "balance_loss_mlp": 1.01769757,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 2.0713530957988597,
      "language_loss": 0.69884419,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.72075123,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.8181629180908203
    },
    {
      "auxiliary_loss_clip": 0.01167459,
      "auxiliary_loss_mlp": 0.01030805,
      "balance_loss_clip": 1.0111028,
      "balance_loss_mlp": 1.0237956,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 1.8445437714809725,
      "language_loss": 0.73281276,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75479537,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.670579433441162
    },
    {
      "auxiliary_loss_clip": 0.0115169,
      "auxiliary_loss_mlp": 0.01025622,
      "balance_loss_clip": 0.93054217,
      "balance_loss_mlp": 1.01849008,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.6197849560003388,
      "language_loss": 0.85778075,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87955391,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.7532198429107666
    },
    {
      "auxiliary_loss_clip": 0.01164016,
      "auxiliary_loss_mlp": 0.01024194,
      "balance_loss_clip": 1.00800824,
      "balance_loss_mlp": 1.01739597,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 1.6757361551916086,
      "language_loss": 0.72706372,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.74894583,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.6112654209136963
    },
    {
      "auxiliary_loss_clip": 0.01156809,
      "auxiliary_loss_mlp": 0.01027609,
      "balance_loss_clip": 0.97174042,
      "balance_loss_mlp": 1.02051032,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 1.980082088596911,
      "language_loss": 0.86461306,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.8864572,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.7662770748138428
    },
    {
      "auxiliary_loss_clip": 0.01160395,
      "auxiliary_loss_mlp": 0.01025421,
      "balance_loss_clip": 0.85672653,
      "balance_loss_mlp": 1.01862311,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.3862560348193242,
      "language_loss": 0.64117467,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.66303283,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.735856294631958
    },
    {
      "auxiliary_loss_clip": 0.01161213,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 0.97073042,
      "balance_loss_mlp": 1.01699877,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.343141249550898,
      "language_loss": 0.85374999,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.87559742,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 2.5920047760009766
    },
    {
      "auxiliary_loss_clip": 0.01157532,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 0.96585274,
      "balance_loss_mlp": 1.01978374,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 2.3936042307545358,
      "language_loss": 0.78837788,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.8102324,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.6600735187530518
    },
    {
      "auxiliary_loss_clip": 0.0115157,
      "auxiliary_loss_mlp": 0.0101921,
      "balance_loss_clip": 1.00345969,
      "balance_loss_mlp": 1.0124929,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 1.658413085795564,
      "language_loss": 0.87408471,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89579254,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 3.45243763923645
    },
    {
      "auxiliary_loss_clip": 0.0115822,
      "auxiliary_loss_mlp": 0.01021316,
      "balance_loss_clip": 1.0063231,
      "balance_loss_mlp": 1.01420486,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 1.5094272285220374,
      "language_loss": 0.70477629,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72657168,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.72623610496521
    },
    {
      "auxiliary_loss_clip": 0.01160848,
      "auxiliary_loss_mlp": 0.01023362,
      "balance_loss_clip": 1.00992227,
      "balance_loss_mlp": 1.01626337,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 1.5484514090439212,
      "language_loss": 0.75735784,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.7791999,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.6268696784973145
    },
    {
      "auxiliary_loss_clip": 0.01165162,
      "auxiliary_loss_mlp": 0.01031988,
      "balance_loss_clip": 0.93168908,
      "balance_loss_mlp": 1.02528524,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 1.8396425809729118,
      "language_loss": 0.7571227,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.77909422,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 2.763134717941284
    },
    {
      "auxiliary_loss_clip": 0.01166818,
      "auxiliary_loss_mlp": 0.01023387,
      "balance_loss_clip": 1.00780654,
      "balance_loss_mlp": 1.01647305,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 2.9694113149261585,
      "language_loss": 0.69297099,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71487308,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 3.6141676902770996
    },
    {
      "auxiliary_loss_clip": 0.01160504,
      "auxiliary_loss_mlp": 0.01022123,
      "balance_loss_clip": 0.8969714,
      "balance_loss_mlp": 1.01542902,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.4879136339524637,
      "language_loss": 0.75640202,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.77822834,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.7304601669311523
    },
    {
      "auxiliary_loss_clip": 0.01160372,
      "auxiliary_loss_mlp": 0.01023613,
      "balance_loss_clip": 1.00603867,
      "balance_loss_mlp": 1.01689231,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 1.7467213625749352,
      "language_loss": 0.71777928,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.73961908,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 3.368116855621338
    },
    {
      "auxiliary_loss_clip": 0.01163295,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.0464325,
      "balance_loss_mlp": 1.01982641,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 2.6574293605647488,
      "language_loss": 0.74637121,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.76826787,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 3.452266216278076
    },
    {
      "auxiliary_loss_clip": 0.01168233,
      "auxiliary_loss_mlp": 0.0112263,
      "balance_loss_clip": 1.01067376,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 2.244232793921065,
      "language_loss": 0.74622333,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.76913196,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.588355779647827
    },
    {
      "auxiliary_loss_clip": 0.01157681,
      "auxiliary_loss_mlp": 0.01029299,
      "balance_loss_clip": 0.96959448,
      "balance_loss_mlp": 1.02185464,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 2.1070092935868105,
      "language_loss": 0.78771526,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.80958509,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.6094188690185547
    },
    {
      "auxiliary_loss_clip": 0.01165463,
      "auxiliary_loss_mlp": 0.01122715,
      "balance_loss_clip": 0.93374515,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 1.8221124288104087,
      "language_loss": 0.78745985,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.81034172,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.6553544998168945
    },
    {
      "auxiliary_loss_clip": 0.01167872,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.007936,
      "balance_loss_mlp": 1.02087331,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 1.885664928381305,
      "language_loss": 0.74559236,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.7675482,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.5658910274505615
    },
    {
      "auxiliary_loss_clip": 0.01161962,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 1.00969982,
      "balance_loss_mlp": 1.01933169,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.827058406375046,
      "language_loss": 0.75305694,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.77493286,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.677183151245117
    },
    {
      "auxiliary_loss_clip": 0.01164821,
      "auxiliary_loss_mlp": 0.01022344,
      "balance_loss_clip": 0.89395082,
      "balance_loss_mlp": 1.01551056,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 5.191015595482234,
      "language_loss": 0.65949678,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68136841,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 2.697948932647705
    },
    {
      "auxiliary_loss_clip": 0.01166786,
      "auxiliary_loss_mlp": 0.01027822,
      "balance_loss_clip": 1.01217997,
      "balance_loss_mlp": 1.0203476,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 2.507926396011683,
      "language_loss": 0.94160497,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.96355104,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 2.6519038677215576
    },
    {
      "auxiliary_loss_clip": 0.01165411,
      "auxiliary_loss_mlp": 0.01027362,
      "balance_loss_clip": 1.04786968,
      "balance_loss_mlp": 1.0206176,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 1.767304068484276,
      "language_loss": 0.84806263,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.86999035,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.6176931858062744
    },
    {
      "auxiliary_loss_clip": 0.011649,
      "auxiliary_loss_mlp": 0.0102378,
      "balance_loss_clip": 1.00822496,
      "balance_loss_mlp": 1.01710987,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 1.7355674564043486,
      "language_loss": 0.75630873,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.7781955,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 2.6536266803741455
    },
    {
      "auxiliary_loss_clip": 0.01167365,
      "auxiliary_loss_mlp": 0.01027366,
      "balance_loss_clip": 1.00905871,
      "balance_loss_mlp": 1.02059758,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 1.8295482857446523,
      "language_loss": 0.70681369,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.72876102,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 2.714266300201416
    },
    {
      "auxiliary_loss_clip": 0.01155242,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 0.92902422,
      "balance_loss_mlp": 1.01824355,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 1.8547026962358886,
      "language_loss": 0.74165285,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.76345897,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.705111026763916
    },
    {
      "auxiliary_loss_clip": 0.01162561,
      "auxiliary_loss_mlp": 0.01025549,
      "balance_loss_clip": 0.97097868,
      "balance_loss_mlp": 1.01836395,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 1.8660396679626041,
      "language_loss": 0.7054773,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.7273584,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.739351511001587
    },
    {
      "auxiliary_loss_clip": 0.01158137,
      "auxiliary_loss_mlp": 0.01026426,
      "balance_loss_clip": 0.96743393,
      "balance_loss_mlp": 1.01877856,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 1.8540250422048077,
      "language_loss": 0.64675564,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.66860127,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.8198585510253906
    },
    {
      "auxiliary_loss_clip": 0.01162386,
      "auxiliary_loss_mlp": 0.01027474,
      "balance_loss_clip": 0.8964864,
      "balance_loss_mlp": 1.01979685,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 1.8538252040698278,
      "language_loss": 0.84078777,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.8626864,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.7609119415283203
    },
    {
      "auxiliary_loss_clip": 0.01165843,
      "auxiliary_loss_mlp": 0.01023587,
      "balance_loss_clip": 1.04656076,
      "balance_loss_mlp": 1.01640141,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 1.9567913001221384,
      "language_loss": 0.75776142,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.77965569,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.5620622634887695
    },
    {
      "auxiliary_loss_clip": 0.01165134,
      "auxiliary_loss_mlp": 0.01025176,
      "balance_loss_clip": 0.93116736,
      "balance_loss_mlp": 1.01772523,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 1.7884689281639656,
      "language_loss": 0.73214412,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75404727,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.6825873851776123
    },
    {
      "auxiliary_loss_clip": 0.01169233,
      "auxiliary_loss_mlp": 0.0103259,
      "balance_loss_clip": 1.01244462,
      "balance_loss_mlp": 1.02555656,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.8141794314171815,
      "language_loss": 0.78248525,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.8045035,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.666520833969116
    },
    {
      "auxiliary_loss_clip": 0.01172853,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 1.04979038,
      "balance_loss_mlp": 1.02039742,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 2.1266328749235743,
      "language_loss": 0.68198824,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.70399868,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 2.648160696029663
    },
    {
      "auxiliary_loss_clip": 0.01165826,
      "auxiliary_loss_mlp": 0.01021308,
      "balance_loss_clip": 0.93222463,
      "balance_loss_mlp": 1.01402378,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 1.8307872216542966,
      "language_loss": 0.7799322,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80180353,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 3.476155996322632
    },
    {
      "auxiliary_loss_clip": 0.01159077,
      "auxiliary_loss_mlp": 0.01028672,
      "balance_loss_clip": 0.97209328,
      "balance_loss_mlp": 1.02126884,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.7493609917968451,
      "language_loss": 0.84915251,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87102997,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 2.6710894107818604
    },
    {
      "auxiliary_loss_clip": 0.01161965,
      "auxiliary_loss_mlp": 0.01025715,
      "balance_loss_clip": 0.89744014,
      "balance_loss_mlp": 1.01875019,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 2.3954847564535626,
      "language_loss": 0.7632519,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78512871,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.7326090335845947
    },
    {
      "auxiliary_loss_clip": 0.01164962,
      "auxiliary_loss_mlp": 0.01122862,
      "balance_loss_clip": 1.01050532,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 4.381879548386204,
      "language_loss": 0.81156605,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.83444428,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.6230576038360596
    },
    {
      "auxiliary_loss_clip": 0.0116113,
      "auxiliary_loss_mlp": 0.01026733,
      "balance_loss_clip": 1.00553262,
      "balance_loss_mlp": 1.01904678,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 1.7612118990162737,
      "language_loss": 0.55727434,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.579153,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 3.7262144088745117
    },
    {
      "auxiliary_loss_clip": 0.01166395,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.04801023,
      "balance_loss_mlp": 1.02227402,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 2.086325788908519,
      "language_loss": 0.83020407,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85216272,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.5785064697265625
    },
    {
      "auxiliary_loss_clip": 0.0116649,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.04888093,
      "balance_loss_mlp": 1.02189422,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 2.022109409372343,
      "language_loss": 0.84211129,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86406863,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 3.562169075012207
    },
    {
      "auxiliary_loss_clip": 0.01167248,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 0.89795136,
      "balance_loss_mlp": 1.01694763,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.142883395119379,
      "language_loss": 0.80965376,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.83156395,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 3.5864689350128174
    },
    {
      "auxiliary_loss_clip": 0.01163774,
      "auxiliary_loss_mlp": 0.01022394,
      "balance_loss_clip": 0.97037232,
      "balance_loss_mlp": 1.01606059,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 1.9502269085259785,
      "language_loss": 0.7957418,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.81760347,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.593554973602295
    },
    {
      "auxiliary_loss_clip": 0.01169359,
      "auxiliary_loss_mlp": 0.01122073,
      "balance_loss_clip": 1.01041436,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.371566994486956,
      "language_loss": 0.58841491,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.6113292,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.579714059829712
    },
    {
      "auxiliary_loss_clip": 0.01168298,
      "auxiliary_loss_mlp": 0.01021299,
      "balance_loss_clip": 0.93217826,
      "balance_loss_mlp": 1.01423264,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 1.637791784959654,
      "language_loss": 0.73501182,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.75690782,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 2.758711576461792
    },
    {
      "auxiliary_loss_clip": 0.01166695,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.01059282,
      "balance_loss_mlp": 1.01958323,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.1574986548034762,
      "language_loss": 0.7043156,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.72624344,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.582824945449829
    },
    {
      "auxiliary_loss_clip": 0.01159874,
      "auxiliary_loss_mlp": 0.01030802,
      "balance_loss_clip": 0.92915118,
      "balance_loss_mlp": 1.02345896,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 1.931899808997864,
      "language_loss": 0.77799594,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.79990268,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 2.622356414794922
    },
    {
      "auxiliary_loss_clip": 0.01168619,
      "auxiliary_loss_mlp": 0.01024006,
      "balance_loss_clip": 0.96886867,
      "balance_loss_mlp": 1.01700521,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 2.3289440175871094,
      "language_loss": 0.73474717,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.75667346,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 2.565768241882324
    },
    {
      "auxiliary_loss_clip": 0.01161856,
      "auxiliary_loss_mlp": 0.01027246,
      "balance_loss_clip": 0.96992892,
      "balance_loss_mlp": 1.02025783,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 1.8884053259697025,
      "language_loss": 0.71876514,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74065614,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 2.627440929412842
    },
    {
      "auxiliary_loss_clip": 0.01167005,
      "auxiliary_loss_mlp": 0.01025097,
      "balance_loss_clip": 1.00714242,
      "balance_loss_mlp": 1.01793826,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 2.5786181427644053,
      "language_loss": 0.74172372,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.7636447,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 2.637256383895874
    },
    {
      "auxiliary_loss_clip": 0.01166577,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.04751766,
      "balance_loss_mlp": 1.01909637,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 1.626235173047815,
      "language_loss": 0.7930662,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.81499797,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.576812744140625
    },
    {
      "auxiliary_loss_clip": 0.01160009,
      "auxiliary_loss_mlp": 0.01025289,
      "balance_loss_clip": 0.93383098,
      "balance_loss_mlp": 1.01889336,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 1.9495436241501116,
      "language_loss": 0.69580704,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.71766007,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 2.6712183952331543
    },
    {
      "auxiliary_loss_clip": 0.01164091,
      "auxiliary_loss_mlp": 0.01025338,
      "balance_loss_clip": 1.04684687,
      "balance_loss_mlp": 1.01846588,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 1.7999505421006932,
      "language_loss": 0.64948261,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.67137694,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 2.6309142112731934
    },
    {
      "auxiliary_loss_clip": 0.0116581,
      "auxiliary_loss_mlp": 0.0102309,
      "balance_loss_clip": 1.01033854,
      "balance_loss_mlp": 1.01565719,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 1.8368746153656172,
      "language_loss": 0.74178857,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.76367754,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.580151081085205
    },
    {
      "auxiliary_loss_clip": 0.01149479,
      "auxiliary_loss_mlp": 0.01025957,
      "balance_loss_clip": 0.93110538,
      "balance_loss_mlp": 1.01922464,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 1.939015198641835,
      "language_loss": 0.69955158,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.72130597,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.705045223236084
    },
    {
      "auxiliary_loss_clip": 0.01167175,
      "auxiliary_loss_mlp": 0.01021051,
      "balance_loss_clip": 1.00934672,
      "balance_loss_mlp": 1.01442325,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 3.132815001756222,
      "language_loss": 0.73672438,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75860655,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.6445887088775635
    },
    {
      "auxiliary_loss_clip": 0.0116465,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.0482372,
      "balance_loss_mlp": 1.02146029,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.4894142355648143,
      "language_loss": 0.79086447,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81279057,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.6047513484954834
    },
    {
      "auxiliary_loss_clip": 0.0116432,
      "auxiliary_loss_mlp": 0.01121402,
      "balance_loss_clip": 0.97146076,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.514165761224923,
      "language_loss": 0.84725845,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.8701157,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 2.6834921836853027
    },
    {
      "auxiliary_loss_clip": 0.01166906,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.00809073,
      "balance_loss_mlp": 1.01909351,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 13.360265846522767,
      "language_loss": 0.75005686,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77198756,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.625354290008545
    },
    {
      "auxiliary_loss_clip": 0.01165346,
      "auxiliary_loss_mlp": 0.01023772,
      "balance_loss_clip": 0.97048223,
      "balance_loss_mlp": 1.01714754,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.7281509420281362,
      "language_loss": 0.66360033,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68549156,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.6122283935546875
    },
    {
      "auxiliary_loss_clip": 0.01156565,
      "auxiliary_loss_mlp": 0.01029278,
      "balance_loss_clip": 0.89425009,
      "balance_loss_mlp": 1.02158308,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 1.7674990430664679,
      "language_loss": 0.78587604,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.80773443,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 2.7441091537475586
    },
    {
      "auxiliary_loss_clip": 0.01167012,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.04805636,
      "balance_loss_mlp": 1.0211997,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 1.8580324882846138,
      "language_loss": 0.79607469,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81802464,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 3.4118142127990723
    },
    {
      "auxiliary_loss_clip": 0.01168551,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.00835037,
      "balance_loss_mlp": 1.01861989,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 2.11332310360186,
      "language_loss": 0.78821814,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81015867,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 2.553889274597168
    },
    {
      "auxiliary_loss_clip": 0.0116499,
      "auxiliary_loss_mlp": 0.01024945,
      "balance_loss_clip": 0.89399302,
      "balance_loss_mlp": 1.01737499,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 1.9308419025420278,
      "language_loss": 0.85424334,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.87614262,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.677792549133301
    },
    {
      "auxiliary_loss_clip": 0.01171552,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.04731011,
      "balance_loss_mlp": 1.01882625,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.261615513204646,
      "language_loss": 0.81767529,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.83965522,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 3.535935401916504
    },
    {
      "auxiliary_loss_clip": 0.0116751,
      "auxiliary_loss_mlp": 0.01022538,
      "balance_loss_clip": 1.04969454,
      "balance_loss_mlp": 1.01545954,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 2.1059827709399794,
      "language_loss": 0.78034478,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.80224526,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 2.5680792331695557
    },
    {
      "auxiliary_loss_clip": 0.01163621,
      "auxiliary_loss_mlp": 0.01023453,
      "balance_loss_clip": 1.04699719,
      "balance_loss_mlp": 1.01674783,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 2.008896282899265,
      "language_loss": 0.72163248,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74350321,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.526439905166626
    },
    {
      "auxiliary_loss_clip": 0.01165978,
      "auxiliary_loss_mlp": 0.01025213,
      "balance_loss_clip": 1.04670906,
      "balance_loss_mlp": 1.01808476,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 1.6305005386775437,
      "language_loss": 0.74112421,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.76303613,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 3.4182393550872803
    },
    {
      "auxiliary_loss_clip": 0.01167449,
      "auxiliary_loss_mlp": 0.01026508,
      "balance_loss_clip": 0.93021125,
      "balance_loss_mlp": 1.01875949,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.176828289628841,
      "language_loss": 0.90256405,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.92450362,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 3.6745169162750244
    },
    {
      "auxiliary_loss_clip": 0.01066567,
      "auxiliary_loss_mlp": 0.01115651,
      "balance_loss_clip": 0.89687729,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.7488886116941545,
      "language_loss": 0.56929874,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.5911209,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.275259494781494
    },
    {
      "auxiliary_loss_clip": 0.01145518,
      "auxiliary_loss_mlp": 0.01122715,
      "balance_loss_clip": 0.85214388,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 1.858715099217798,
      "language_loss": 0.79953575,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.82221812,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 3.032892942428589
    },
    {
      "auxiliary_loss_clip": 0.01169954,
      "auxiliary_loss_mlp": 0.01024049,
      "balance_loss_clip": 1.01147985,
      "balance_loss_mlp": 1.01736462,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.8567855523981067,
      "language_loss": 0.73353958,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.75547957,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.7060327529907227
    },
    {
      "auxiliary_loss_clip": 0.01155249,
      "auxiliary_loss_mlp": 0.01122453,
      "balance_loss_clip": 0.89106578,
      "balance_loss_mlp": 0.0,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.1183488319457253,
      "language_loss": 0.70596498,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.728742,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 2.7141165733337402
    },
    {
      "auxiliary_loss_clip": 0.01151742,
      "auxiliary_loss_mlp": 0.0102395,
      "balance_loss_clip": 0.93088758,
      "balance_loss_mlp": 1.01668429,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 1.807601038257503,
      "language_loss": 0.75598794,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77774489,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 2.6928350925445557
    },
    {
      "auxiliary_loss_clip": 0.01164867,
      "auxiliary_loss_mlp": 0.01021239,
      "balance_loss_clip": 1.04597056,
      "balance_loss_mlp": 1.01426554,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 2.3282610611213523,
      "language_loss": 0.81774342,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.83960444,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.5746395587921143
    },
    {
      "auxiliary_loss_clip": 0.01155761,
      "auxiliary_loss_mlp": 0.01025524,
      "balance_loss_clip": 0.93208635,
      "balance_loss_mlp": 1.01867557,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 1.9490531269248434,
      "language_loss": 0.83791929,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.85973215,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 2.705136299133301
    },
    {
      "auxiliary_loss_clip": 0.01166454,
      "auxiliary_loss_mlp": 0.0102467,
      "balance_loss_clip": 1.011693,
      "balance_loss_mlp": 1.01783371,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 2.1125260073390915,
      "language_loss": 0.75330108,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.77521229,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 2.6027979850769043
    },
    {
      "auxiliary_loss_clip": 0.01169522,
      "auxiliary_loss_mlp": 0.01026877,
      "balance_loss_clip": 1.00932074,
      "balance_loss_mlp": 1.02011788,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 1.820358253673094,
      "language_loss": 0.61509103,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.63705504,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.600740432739258
    },
    {
      "auxiliary_loss_clip": 0.01173094,
      "auxiliary_loss_mlp": 0.01027091,
      "balance_loss_clip": 1.01083863,
      "balance_loss_mlp": 1.01900303,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 1.8423519918909608,
      "language_loss": 0.76598495,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78798676,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 2.6336188316345215
    },
    {
      "auxiliary_loss_clip": 0.01167473,
      "auxiliary_loss_mlp": 0.01030564,
      "balance_loss_clip": 1.00898004,
      "balance_loss_mlp": 1.02327991,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 1.9126704990968526,
      "language_loss": 0.80579495,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.82777524,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.645183801651001
    },
    {
      "auxiliary_loss_clip": 0.01161034,
      "auxiliary_loss_mlp": 0.0102496,
      "balance_loss_clip": 0.96802348,
      "balance_loss_mlp": 1.01836765,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 1.677925309942129,
      "language_loss": 0.77471733,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79657722,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.70439076423645
    },
    {
      "auxiliary_loss_clip": 0.01165578,
      "auxiliary_loss_mlp": 0.01022523,
      "balance_loss_clip": 1.01143193,
      "balance_loss_mlp": 1.01528144,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.6818766955815936,
      "language_loss": 0.5958488,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.61772978,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.7960524559020996
    },
    {
      "auxiliary_loss_clip": 0.01163325,
      "auxiliary_loss_mlp": 0.01023957,
      "balance_loss_clip": 1.01019979,
      "balance_loss_mlp": 1.0171622,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 1.70425530704488,
      "language_loss": 0.61660206,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.63847494,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.7842600345611572
    },
    {
      "auxiliary_loss_clip": 0.01162088,
      "auxiliary_loss_mlp": 0.0102483,
      "balance_loss_clip": 0.97058809,
      "balance_loss_mlp": 1.01751077,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 1.885667938191396,
      "language_loss": 0.74450582,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.76637501,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.7045373916625977
    },
    {
      "auxiliary_loss_clip": 0.01157627,
      "auxiliary_loss_mlp": 0.01028148,
      "balance_loss_clip": 0.96958596,
      "balance_loss_mlp": 1.02138567,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 1.7409234299654046,
      "language_loss": 0.75307262,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77493036,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.7672131061553955
    },
    {
      "auxiliary_loss_clip": 0.01152227,
      "auxiliary_loss_mlp": 0.01022856,
      "balance_loss_clip": 1.00580955,
      "balance_loss_mlp": 1.01620972,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.5250902144063092,
      "language_loss": 0.84913248,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87088329,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.658827543258667
    },
    {
      "auxiliary_loss_clip": 0.01155533,
      "auxiliary_loss_mlp": 0.01023457,
      "balance_loss_clip": 0.93194735,
      "balance_loss_mlp": 1.01583338,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 1.5454503212304773,
      "language_loss": 0.63620603,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.65799594,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 2.848465919494629
    },
    {
      "auxiliary_loss_clip": 0.01169161,
      "auxiliary_loss_mlp": 0.01029778,
      "balance_loss_clip": 0.93505573,
      "balance_loss_mlp": 1.02234852,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 1.7752231932042326,
      "language_loss": 0.71231401,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73430347,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 3.5963668823242188
    },
    {
      "auxiliary_loss_clip": 0.01165292,
      "auxiliary_loss_mlp": 0.01025472,
      "balance_loss_clip": 1.0473628,
      "balance_loss_mlp": 1.0186559,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 1.8965066248052658,
      "language_loss": 0.796785,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.81869268,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 2.5923895835876465
    },
    {
      "auxiliary_loss_clip": 0.01166422,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 0.93334079,
      "balance_loss_mlp": 1.0202471,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.4014222093469142,
      "language_loss": 0.8093729,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.83131123,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.8650455474853516
    },
    {
      "auxiliary_loss_clip": 0.01164945,
      "auxiliary_loss_mlp": 0.01028704,
      "balance_loss_clip": 0.97275436,
      "balance_loss_mlp": 1.0217247,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 5.6145772236903335,
      "language_loss": 0.79193056,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.81386703,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 3.553619384765625
    },
    {
      "auxiliary_loss_clip": 0.01151171,
      "auxiliary_loss_mlp": 0.0102126,
      "balance_loss_clip": 0.93106848,
      "balance_loss_mlp": 1.01498389,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 1.781912824474208,
      "language_loss": 0.82534683,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84707117,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 2.730072021484375
    },
    {
      "auxiliary_loss_clip": 0.01160347,
      "auxiliary_loss_mlp": 0.01024361,
      "balance_loss_clip": 0.97179925,
      "balance_loss_mlp": 1.01734865,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 2.3876711174565015,
      "language_loss": 0.76452643,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.7863735,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.6385321617126465
    },
    {
      "auxiliary_loss_clip": 0.01164658,
      "auxiliary_loss_mlp": 0.01030106,
      "balance_loss_clip": 1.00857472,
      "balance_loss_mlp": 1.02274227,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 3.3805610783369904,
      "language_loss": 0.6936487,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71559632,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 3.6478829383850098
    },
    {
      "auxiliary_loss_clip": 0.01171597,
      "auxiliary_loss_mlp": 0.01032508,
      "balance_loss_clip": 1.04976106,
      "balance_loss_mlp": 1.02503371,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 5.612931447965669,
      "language_loss": 0.73322314,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.75526422,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 3.521261215209961
    },
    {
      "auxiliary_loss_clip": 0.01063338,
      "auxiliary_loss_mlp": 0.01003469,
      "balance_loss_clip": 0.97130734,
      "balance_loss_mlp": 1.0018481,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8331057262060332,
      "language_loss": 0.56680202,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58747011,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.140091896057129
    },
    {
      "auxiliary_loss_clip": 0.01164508,
      "auxiliary_loss_mlp": 0.01022486,
      "balance_loss_clip": 1.04709983,
      "balance_loss_mlp": 1.01530957,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 2.0759906072106173,
      "language_loss": 0.85317445,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87504435,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 2.6479554176330566
    },
    {
      "auxiliary_loss_clip": 0.01169221,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.01058269,
      "balance_loss_mlp": 1.01773584,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 1.5526746079083533,
      "language_loss": 0.76740319,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78934705,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.6288821697235107
    },
    {
      "auxiliary_loss_clip": 0.01171587,
      "auxiliary_loss_mlp": 0.01031805,
      "balance_loss_clip": 1.04907882,
      "balance_loss_mlp": 1.02334702,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 1.8109594424633098,
      "language_loss": 0.8184166,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84045053,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.6517839431762695
    },
    {
      "auxiliary_loss_clip": 0.01161461,
      "auxiliary_loss_mlp": 0.01020055,
      "balance_loss_clip": 0.96637523,
      "balance_loss_mlp": 1.01342142,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 1.9480878904356338,
      "language_loss": 0.81989563,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84171081,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 2.706088066101074
    },
    {
      "auxiliary_loss_clip": 0.01164764,
      "auxiliary_loss_mlp": 0.01028126,
      "balance_loss_clip": 0.97235036,
      "balance_loss_mlp": 1.02115846,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 2.5975539653435415,
      "language_loss": 0.80650496,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.82843381,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 2.6546168327331543
    },
    {
      "auxiliary_loss_clip": 0.01151422,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 0.93055129,
      "balance_loss_mlp": 1.0206356,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.5600632737872202,
      "language_loss": 0.77142084,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.79321331,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.72135853767395
    },
    {
      "auxiliary_loss_clip": 0.01171548,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 0.97236729,
      "balance_loss_mlp": 1.01986933,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 5.4577111321395755,
      "language_loss": 0.80632341,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.82831079,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 2.6490442752838135
    },
    {
      "auxiliary_loss_clip": 0.01163539,
      "auxiliary_loss_mlp": 0.01026675,
      "balance_loss_clip": 0.97020823,
      "balance_loss_mlp": 1.01999044,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.700513633073312,
      "language_loss": 0.61028397,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63218606,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 2.815195083618164
    },
    {
      "auxiliary_loss_clip": 0.01162623,
      "auxiliary_loss_mlp": 0.0112203,
      "balance_loss_clip": 0.93626738,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.776752030748092,
      "language_loss": 0.80174887,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.82459539,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 2.7090578079223633
    },
    {
      "auxiliary_loss_clip": 0.01159458,
      "auxiliary_loss_mlp": 0.01024307,
      "balance_loss_clip": 0.97131848,
      "balance_loss_mlp": 1.01709485,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 2.373230289191325,
      "language_loss": 0.73404706,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75588471,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.700500011444092
    },
    {
      "auxiliary_loss_clip": 0.01165126,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 0.97004783,
      "balance_loss_mlp": 1.01846957,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.6797076077967725,
      "language_loss": 0.82111609,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84302723,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.6684625148773193
    },
    {
      "auxiliary_loss_clip": 0.01164903,
      "auxiliary_loss_mlp": 0.01027258,
      "balance_loss_clip": 1.00912881,
      "balance_loss_mlp": 1.02057981,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.379451553379144,
      "language_loss": 0.76506531,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78698695,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.7119174003601074
    },
    {
      "auxiliary_loss_clip": 0.01167439,
      "auxiliary_loss_mlp": 0.01024766,
      "balance_loss_clip": 1.04711628,
      "balance_loss_mlp": 1.01755428,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 2.6509743745280274,
      "language_loss": 0.73999429,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76191634,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.669278860092163
    },
    {
      "auxiliary_loss_clip": 0.01161967,
      "auxiliary_loss_mlp": 0.01030329,
      "balance_loss_clip": 0.89459705,
      "balance_loss_mlp": 1.02241397,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 1.9829536118873539,
      "language_loss": 0.73387742,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75580037,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.740239143371582
    },
    {
      "auxiliary_loss_clip": 0.01161805,
      "auxiliary_loss_mlp": 0.01022595,
      "balance_loss_clip": 0.9330042,
      "balance_loss_mlp": 1.01528716,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 2.3013481445528865,
      "language_loss": 0.73446363,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.7563076,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.6460049152374268
    },
    {
      "auxiliary_loss_clip": 0.01073229,
      "auxiliary_loss_mlp": 0.01000653,
      "balance_loss_clip": 0.82366753,
      "balance_loss_mlp": 0.99902022,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.85302078784982,
      "language_loss": 0.57502127,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59576011,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.3624746799468994
    },
    {
      "auxiliary_loss_clip": 0.01154781,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 0.85269904,
      "balance_loss_mlp": 1.0174253,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 1.5781640397206365,
      "language_loss": 0.7943505,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.81614029,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 3.885612726211548
    },
    {
      "auxiliary_loss_clip": 0.01164795,
      "auxiliary_loss_mlp": 0.01030278,
      "balance_loss_clip": 0.89370382,
      "balance_loss_mlp": 1.02315784,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 1.6468334475865711,
      "language_loss": 0.72164887,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74359959,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 2.742582321166992
    },
    {
      "auxiliary_loss_clip": 0.01156852,
      "auxiliary_loss_mlp": 0.01026127,
      "balance_loss_clip": 0.9320246,
      "balance_loss_mlp": 1.01880801,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 1.8975994760199184,
      "language_loss": 0.8041507,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82598054,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 2.7576613426208496
    },
    {
      "auxiliary_loss_clip": 0.01166011,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.00852668,
      "balance_loss_mlp": 1.01935256,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 4.537728434820711,
      "language_loss": 0.76001155,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.78193545,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.6696321964263916
    },
    {
      "auxiliary_loss_clip": 0.01165334,
      "auxiliary_loss_mlp": 0.01020052,
      "balance_loss_clip": 1.04915714,
      "balance_loss_mlp": 1.01399565,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 2.0015343213832226,
      "language_loss": 0.74887675,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.77073061,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 3.51523756980896
    },
    {
      "auxiliary_loss_clip": 0.01164524,
      "auxiliary_loss_mlp": 0.01023764,
      "balance_loss_clip": 1.00775266,
      "balance_loss_mlp": 1.01675749,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 3.5374358331956874,
      "language_loss": 0.77167761,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.7935605,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.662196636199951
    },
    {
      "auxiliary_loss_clip": 0.01168409,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.01069438,
      "balance_loss_mlp": 1.01829433,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 3.289768703205141,
      "language_loss": 0.71877444,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.7407195,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 2.755389451980591
    },
    {
      "auxiliary_loss_clip": 0.01156217,
      "auxiliary_loss_mlp": 0.01023408,
      "balance_loss_clip": 1.00633478,
      "balance_loss_mlp": 1.01672673,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 3.1663892180787268,
      "language_loss": 0.71070588,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.73250216,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 3.4798786640167236
    },
    {
      "auxiliary_loss_clip": 0.01169133,
      "auxiliary_loss_mlp": 0.01027759,
      "balance_loss_clip": 1.0483954,
      "balance_loss_mlp": 1.02018952,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 9.8572257836287,
      "language_loss": 0.71588999,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73785889,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 3.590686798095703
    },
    {
      "auxiliary_loss_clip": 0.01163017,
      "auxiliary_loss_mlp": 0.010232,
      "balance_loss_clip": 0.97281969,
      "balance_loss_mlp": 1.01654267,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 1.3912701231233744,
      "language_loss": 0.68332541,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70518756,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 2.773472785949707
    },
    {
      "auxiliary_loss_clip": 0.01169286,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.0119319,
      "balance_loss_mlp": 1.02089524,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 2.0395478423559807,
      "language_loss": 0.79989612,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.82187265,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 2.701533794403076
    },
    {
      "auxiliary_loss_clip": 0.01159402,
      "auxiliary_loss_mlp": 0.01022203,
      "balance_loss_clip": 0.93122119,
      "balance_loss_mlp": 1.01479375,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 5.188810080607118,
      "language_loss": 0.75117719,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77299321,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.6558244228363037
    },
    {
      "auxiliary_loss_clip": 0.01166195,
      "auxiliary_loss_mlp": 0.01122297,
      "balance_loss_clip": 1.04784846,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 1.8056718624267027,
      "language_loss": 0.70942259,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.73230749,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.6373279094696045
    },
    {
      "auxiliary_loss_clip": 0.01166148,
      "auxiliary_loss_mlp": 0.01023967,
      "balance_loss_clip": 1.00759459,
      "balance_loss_mlp": 1.01728272,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 2.279887704379277,
      "language_loss": 0.61200005,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.63390124,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 2.610978841781616
    },
    {
      "auxiliary_loss_clip": 0.01163485,
      "auxiliary_loss_mlp": 0.01025017,
      "balance_loss_clip": 0.93041217,
      "balance_loss_mlp": 1.01741791,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 2.3627618940188952,
      "language_loss": 0.82398367,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.84586871,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.678744077682495
    },
    {
      "auxiliary_loss_clip": 0.01161845,
      "auxiliary_loss_mlp": 0.01022246,
      "balance_loss_clip": 0.97087693,
      "balance_loss_mlp": 1.01518297,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.1186791690898374,
      "language_loss": 0.81147897,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.8333199,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 2.754265785217285
    },
    {
      "auxiliary_loss_clip": 0.01166932,
      "auxiliary_loss_mlp": 0.01023659,
      "balance_loss_clip": 1.04827547,
      "balance_loss_mlp": 1.01658356,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.624104296373448,
      "language_loss": 0.77179492,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79370081,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.604515790939331
    },
    {
      "auxiliary_loss_clip": 0.01159281,
      "auxiliary_loss_mlp": 0.01122275,
      "balance_loss_clip": 0.96741462,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 2.884856457003942,
      "language_loss": 0.7045849,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.72740048,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 2.678764581680298
    },
    {
      "auxiliary_loss_clip": 0.01165679,
      "auxiliary_loss_mlp": 0.01022644,
      "balance_loss_clip": 0.96771532,
      "balance_loss_mlp": 1.01633191,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.592898595161629,
      "language_loss": 0.82787019,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.84975338,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 2.6569571495056152
    },
    {
      "auxiliary_loss_clip": 0.011676,
      "auxiliary_loss_mlp": 0.01023729,
      "balance_loss_clip": 0.96866608,
      "balance_loss_mlp": 1.01644528,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 2.0950630219230373,
      "language_loss": 0.82341224,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.84532547,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.6353065967559814
    },
    {
      "auxiliary_loss_clip": 0.01161937,
      "auxiliary_loss_mlp": 0.01022143,
      "balance_loss_clip": 0.97077608,
      "balance_loss_mlp": 1.01499379,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 1.8573186636696946,
      "language_loss": 0.71778029,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.73962104,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.6455271244049072
    },
    {
      "auxiliary_loss_clip": 0.01159956,
      "auxiliary_loss_mlp": 0.01023626,
      "balance_loss_clip": 0.97026247,
      "balance_loss_mlp": 1.01707816,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 2.0451136315433156,
      "language_loss": 0.72444165,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.74627751,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.658169746398926
    },
    {
      "auxiliary_loss_clip": 0.01162184,
      "auxiliary_loss_mlp": 0.01023382,
      "balance_loss_clip": 0.96892977,
      "balance_loss_mlp": 1.01725125,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.854578105163528,
      "language_loss": 0.79517341,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.81702906,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.6627824306488037
    },
    {
      "auxiliary_loss_clip": 0.01165275,
      "auxiliary_loss_mlp": 0.01026462,
      "balance_loss_clip": 0.96966493,
      "balance_loss_mlp": 1.01917863,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 2.3782382789236696,
      "language_loss": 0.73500443,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.75692177,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.909970998764038
    },
    {
      "auxiliary_loss_clip": 0.01169594,
      "auxiliary_loss_mlp": 0.01033298,
      "balance_loss_clip": 1.00987506,
      "balance_loss_mlp": 1.02543592,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 2.1529485043164667,
      "language_loss": 0.68042028,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.7024492,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.650679111480713
    },
    {
      "auxiliary_loss_clip": 0.01164478,
      "auxiliary_loss_mlp": 0.01023388,
      "balance_loss_clip": 1.04506695,
      "balance_loss_mlp": 1.01643801,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 2.9025538447655115,
      "language_loss": 0.6858933,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.70777202,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 2.611185073852539
    },
    {
      "auxiliary_loss_clip": 0.01173046,
      "auxiliary_loss_mlp": 0.01026544,
      "balance_loss_clip": 0.9748553,
      "balance_loss_mlp": 1.01939154,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 1.9039446417256383,
      "language_loss": 0.78822184,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.81021774,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 3.7174999713897705
    },
    {
      "auxiliary_loss_clip": 0.01166371,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 0.96898961,
      "balance_loss_mlp": 1.02078307,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 2.5722480741919833,
      "language_loss": 0.65569049,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.67763889,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 2.705104351043701
    },
    {
      "auxiliary_loss_clip": 0.01171274,
      "auxiliary_loss_mlp": 0.0103107,
      "balance_loss_clip": 1.01220429,
      "balance_loss_mlp": 1.02391458,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 3.6009478566309014,
      "language_loss": 0.68320775,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.70523113,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 2.706979990005493
    },
    {
      "auxiliary_loss_clip": 0.01155635,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.00629854,
      "balance_loss_mlp": 1.01880097,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 1.9564085218463758,
      "language_loss": 0.69820774,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.7200253,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.677022695541382
    },
    {
      "auxiliary_loss_clip": 0.01170895,
      "auxiliary_loss_mlp": 0.01122546,
      "balance_loss_clip": 1.01045847,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 2.1483704339223113,
      "language_loss": 0.8244921,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84742647,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 3.6923766136169434
    },
    {
      "auxiliary_loss_clip": 0.01166904,
      "auxiliary_loss_mlp": 0.0102424,
      "balance_loss_clip": 1.00864792,
      "balance_loss_mlp": 1.01678038,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 1.9573017671744903,
      "language_loss": 0.85777009,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87968147,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.6717872619628906
    },
    {
      "auxiliary_loss_clip": 0.01157382,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 0.97033703,
      "balance_loss_mlp": 1.02001715,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.423424056706476,
      "language_loss": 0.75974298,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78158045,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 2.656130790710449
    },
    {
      "auxiliary_loss_clip": 0.01166917,
      "auxiliary_loss_mlp": 0.01028965,
      "balance_loss_clip": 0.89678776,
      "balance_loss_mlp": 1.02172899,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 1.8150104959406705,
      "language_loss": 0.80011535,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.82207417,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 3.689608335494995
    },
    {
      "auxiliary_loss_clip": 0.01155696,
      "auxiliary_loss_mlp": 0.01028884,
      "balance_loss_clip": 0.96722227,
      "balance_loss_mlp": 1.02191651,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 1.6502646387991144,
      "language_loss": 0.75631011,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.77815586,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 3.6418304443359375
    },
    {
      "auxiliary_loss_clip": 0.01165294,
      "auxiliary_loss_mlp": 0.01122576,
      "balance_loss_clip": 1.00916266,
      "balance_loss_mlp": 0.0,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.1141380281973587,
      "language_loss": 0.73380989,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.75668859,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.6105470657348633
    },
    {
      "auxiliary_loss_clip": 0.01168611,
      "auxiliary_loss_mlp": 0.01029722,
      "balance_loss_clip": 0.97314161,
      "balance_loss_mlp": 1.0219171,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.6089775617632485,
      "language_loss": 0.7181288,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.74011213,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 2.7258429527282715
    },
    {
      "auxiliary_loss_clip": 0.0106041,
      "auxiliary_loss_mlp": 0.01004157,
      "balance_loss_clip": 0.97359419,
      "balance_loss_mlp": 1.00251234,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7445046187974923,
      "language_loss": 0.54843557,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56908119,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.2516067028045654
    },
    {
      "auxiliary_loss_clip": 0.01163298,
      "auxiliary_loss_mlp": 0.01018207,
      "balance_loss_clip": 1.00957799,
      "balance_loss_mlp": 1.0113734,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 1.993761841722228,
      "language_loss": 0.71122473,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.7330398,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 2.6401901245117188
    },
    {
      "auxiliary_loss_clip": 0.0116565,
      "auxiliary_loss_mlp": 0.0102108,
      "balance_loss_clip": 0.96999007,
      "balance_loss_mlp": 1.01400518,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 1.6692203339012017,
      "language_loss": 0.70605284,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.72792017,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 2.7388386726379395
    },
    {
      "auxiliary_loss_clip": 0.01166644,
      "auxiliary_loss_mlp": 0.01027384,
      "balance_loss_clip": 1.04883599,
      "balance_loss_mlp": 1.02050614,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.7800910707335593,
      "language_loss": 0.76867306,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79061341,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.6355669498443604
    },
    {
      "auxiliary_loss_clip": 0.0116534,
      "auxiliary_loss_mlp": 0.01024443,
      "balance_loss_clip": 0.972736,
      "balance_loss_mlp": 1.01698971,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 1.9347420452300272,
      "language_loss": 0.83532828,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.85722607,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 2.6047346591949463
    },
    {
      "auxiliary_loss_clip": 0.01171586,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 0.97194183,
      "balance_loss_mlp": 1.01896894,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 2.173015319187495,
      "language_loss": 0.62947041,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.65145713,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.6522374153137207
    },
    {
      "auxiliary_loss_clip": 0.01167446,
      "auxiliary_loss_mlp": 0.01022665,
      "balance_loss_clip": 1.0089438,
      "balance_loss_mlp": 1.01529217,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 1.8763501070955628,
      "language_loss": 0.69406915,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.71597028,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 2.7142250537872314
    },
    {
      "auxiliary_loss_clip": 0.01167242,
      "auxiliary_loss_mlp": 0.01030127,
      "balance_loss_clip": 0.96915406,
      "balance_loss_mlp": 1.02309942,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.177480668168622,
      "language_loss": 0.78331792,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.80529165,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 2.632478952407837
    },
    {
      "auxiliary_loss_clip": 0.0116445,
      "auxiliary_loss_mlp": 0.0102637,
      "balance_loss_clip": 0.89496017,
      "balance_loss_mlp": 1.0185616,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.7058087659517354,
      "language_loss": 0.80740982,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.82931805,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.6603939533233643
    },
    {
      "auxiliary_loss_clip": 0.01170034,
      "auxiliary_loss_mlp": 0.01030738,
      "balance_loss_clip": 1.04929137,
      "balance_loss_mlp": 1.02283466,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 1.9718712049293436,
      "language_loss": 0.77865279,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.80066049,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.556337833404541
    },
    {
      "auxiliary_loss_clip": 0.01169926,
      "auxiliary_loss_mlp": 0.01025092,
      "balance_loss_clip": 1.00984955,
      "balance_loss_mlp": 1.01779008,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 1.905701323604986,
      "language_loss": 0.6965214,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.71847153,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.6018354892730713
    },
    {
      "auxiliary_loss_clip": 0.01165042,
      "auxiliary_loss_mlp": 0.01028668,
      "balance_loss_clip": 0.93231708,
      "balance_loss_mlp": 1.02127123,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 3.0935386595049272,
      "language_loss": 0.7837804,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.80571747,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.6690006256103516
    },
    {
      "auxiliary_loss_clip": 0.01165696,
      "auxiliary_loss_mlp": 0.01027299,
      "balance_loss_clip": 0.97110325,
      "balance_loss_mlp": 1.02010822,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 3.094894862898306,
      "language_loss": 0.83061194,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.85254192,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.7193872928619385
    },
    {
      "auxiliary_loss_clip": 0.01172092,
      "auxiliary_loss_mlp": 0.0102941,
      "balance_loss_clip": 0.97204566,
      "balance_loss_mlp": 1.02207232,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.5893325547310937,
      "language_loss": 0.70728511,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.72930014,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.6295359134674072
    },
    {
      "auxiliary_loss_clip": 0.01168068,
      "auxiliary_loss_mlp": 0.01026777,
      "balance_loss_clip": 1.0490303,
      "balance_loss_mlp": 1.02002978,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 1.9620953528870875,
      "language_loss": 0.84501326,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86696172,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 2.62461519241333
    },
    {
      "auxiliary_loss_clip": 0.01169089,
      "auxiliary_loss_mlp": 0.0102349,
      "balance_loss_clip": 0.93411219,
      "balance_loss_mlp": 1.01623356,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 2.1785322861064484,
      "language_loss": 0.85400552,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.87593132,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 3.538027286529541
    },
    {
      "auxiliary_loss_clip": 0.01169501,
      "auxiliary_loss_mlp": 0.01026641,
      "balance_loss_clip": 0.97168612,
      "balance_loss_mlp": 1.0191015,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 1.8952423874824011,
      "language_loss": 0.77989089,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.80185235,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 2.595660448074341
    },
    {
      "auxiliary_loss_clip": 0.01162834,
      "auxiliary_loss_mlp": 0.01023965,
      "balance_loss_clip": 1.006248,
      "balance_loss_mlp": 1.01709819,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 2.192971506056647,
      "language_loss": 0.66326189,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.68512988,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 2.625572919845581
    },
    {
      "auxiliary_loss_clip": 0.01059059,
      "auxiliary_loss_mlp": 0.01003371,
      "balance_loss_clip": 1.00827122,
      "balance_loss_mlp": 1.00174928,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 1.0082626576208837,
      "language_loss": 0.57770741,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59833181,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.2983877658843994
    },
    {
      "auxiliary_loss_clip": 0.01155448,
      "auxiliary_loss_mlp": 0.0103164,
      "balance_loss_clip": 0.9698956,
      "balance_loss_mlp": 1.02428138,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.3336618491557966,
      "language_loss": 0.85273063,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.87460148,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 3.5673534870147705
    },
    {
      "auxiliary_loss_clip": 0.01155699,
      "auxiliary_loss_mlp": 0.01023444,
      "balance_loss_clip": 0.96962142,
      "balance_loss_mlp": 1.01640821,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.6990030885754828,
      "language_loss": 0.79969418,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82148564,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.6703317165374756
    },
    {
      "auxiliary_loss_clip": 0.01161012,
      "auxiliary_loss_mlp": 0.01028739,
      "balance_loss_clip": 0.97204864,
      "balance_loss_mlp": 1.02137756,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 1.6998026706673823,
      "language_loss": 0.75185055,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.77374804,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.7261574268341064
    },
    {
      "auxiliary_loss_clip": 0.01159118,
      "auxiliary_loss_mlp": 0.01026808,
      "balance_loss_clip": 0.93068564,
      "balance_loss_mlp": 1.02001953,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.6740586505343502,
      "language_loss": 0.7947762,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81663549,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 3.597792387008667
    },
    {
      "auxiliary_loss_clip": 0.01167315,
      "auxiliary_loss_mlp": 0.01026636,
      "balance_loss_clip": 1.04728925,
      "balance_loss_mlp": 1.01953435,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.6923867347034802,
      "language_loss": 0.71770388,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.73964334,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 3.5722134113311768
    },
    {
      "auxiliary_loss_clip": 0.01063019,
      "auxiliary_loss_mlp": 0.01003743,
      "balance_loss_clip": 0.93469489,
      "balance_loss_mlp": 1.0020386,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8311551438961264,
      "language_loss": 0.55663514,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57730275,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 3.1225790977478027
    },
    {
      "auxiliary_loss_clip": 0.01167468,
      "auxiliary_loss_mlp": 0.01024771,
      "balance_loss_clip": 0.93335634,
      "balance_loss_mlp": 1.01754105,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 1.743363290953299,
      "language_loss": 0.74089217,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.76281458,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 2.7406373023986816
    },
    {
      "auxiliary_loss_clip": 0.01166042,
      "auxiliary_loss_mlp": 0.01122291,
      "balance_loss_clip": 0.96979463,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.441804880430036,
      "language_loss": 0.68941474,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.71229804,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 2.6964457035064697
    },
    {
      "auxiliary_loss_clip": 0.01163431,
      "auxiliary_loss_mlp": 0.01030224,
      "balance_loss_clip": 0.96948355,
      "balance_loss_mlp": 1.02348602,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 2.029064936768715,
      "language_loss": 0.78892887,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.8108654,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.6506357192993164
    },
    {
      "auxiliary_loss_clip": 0.01167226,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.04753745,
      "balance_loss_mlp": 1.0175885,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 2.777956186516845,
      "language_loss": 0.6767841,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.6986984,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 2.756197690963745
    },
    {
      "auxiliary_loss_clip": 0.01169056,
      "auxiliary_loss_mlp": 0.01027866,
      "balance_loss_clip": 1.01183105,
      "balance_loss_mlp": 1.02070093,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 2.162210389344314,
      "language_loss": 0.82024181,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84221095,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.617975950241089
    },
    {
      "auxiliary_loss_clip": 0.01165163,
      "auxiliary_loss_mlp": 0.01026574,
      "balance_loss_clip": 1.04859412,
      "balance_loss_mlp": 1.01961184,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.7026169087434069,
      "language_loss": 0.79172975,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81364715,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.596271514892578
    },
    {
      "auxiliary_loss_clip": 0.01062665,
      "auxiliary_loss_mlp": 0.010019,
      "balance_loss_clip": 0.97163498,
      "balance_loss_mlp": 1.00023127,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.8791844046577423,
      "language_loss": 0.63107306,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.65171868,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.317721128463745
    },
    {
      "auxiliary_loss_clip": 0.01166098,
      "auxiliary_loss_mlp": 0.01023605,
      "balance_loss_clip": 0.97262025,
      "balance_loss_mlp": 1.01662493,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 2.1260512569632226,
      "language_loss": 0.74600232,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76789927,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.582359552383423
    },
    {
      "auxiliary_loss_clip": 0.01163864,
      "auxiliary_loss_mlp": 0.01031978,
      "balance_loss_clip": 0.96944231,
      "balance_loss_mlp": 1.02414,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 2.50957176234986,
      "language_loss": 0.72105229,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.7430107,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 2.737253189086914
    },
    {
      "auxiliary_loss_clip": 0.01167416,
      "auxiliary_loss_mlp": 0.01026469,
      "balance_loss_clip": 0.97305429,
      "balance_loss_mlp": 1.02008522,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.6689923702430685,
      "language_loss": 0.71996868,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74190748,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.6370906829833984
    },
    {
      "auxiliary_loss_clip": 0.0115502,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 0.85326302,
      "balance_loss_mlp": 1.02066803,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 2.081170329479492,
      "language_loss": 0.71567333,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.73750657,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.8005599975585938
    },
    {
      "auxiliary_loss_clip": 0.01168223,
      "auxiliary_loss_mlp": 0.01030078,
      "balance_loss_clip": 1.00964999,
      "balance_loss_mlp": 1.02256799,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 1.7161279501640767,
      "language_loss": 0.69162548,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.7136085,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.6142873764038086
    },
    {
      "auxiliary_loss_clip": 0.01155859,
      "auxiliary_loss_mlp": 0.01023679,
      "balance_loss_clip": 0.85565215,
      "balance_loss_mlp": 1.01691425,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 1.9330938704181793,
      "language_loss": 0.65822923,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.68002468,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.9257051944732666
    },
    {
      "auxiliary_loss_clip": 0.01171385,
      "auxiliary_loss_mlp": 0.01022667,
      "balance_loss_clip": 1.01164806,
      "balance_loss_mlp": 1.01549673,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 2.0844028199762974,
      "language_loss": 0.74557137,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.76751196,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.7666807174682617
    },
    {
      "auxiliary_loss_clip": 0.01163564,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.04532361,
      "balance_loss_mlp": 1.01765037,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.698420325543636,
      "language_loss": 0.78496581,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80684763,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.5838377475738525
    },
    {
      "auxiliary_loss_clip": 0.01161223,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 0.93378168,
      "balance_loss_mlp": 1.01777375,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 2.6295084745119373,
      "language_loss": 0.69268823,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71454668,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.6616365909576416
    },
    {
      "auxiliary_loss_clip": 0.0116624,
      "auxiliary_loss_mlp": 0.01027997,
      "balance_loss_clip": 1.00923896,
      "balance_loss_mlp": 1.02045667,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 2.2738059769816448,
      "language_loss": 0.76299524,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.78493762,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 3.6969480514526367
    },
    {
      "auxiliary_loss_clip": 0.01162724,
      "auxiliary_loss_mlp": 0.01021877,
      "balance_loss_clip": 0.96911246,
      "balance_loss_mlp": 1.01467121,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 1.9076483983222132,
      "language_loss": 0.82876289,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.85060894,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.724283218383789
    },
    {
      "auxiliary_loss_clip": 0.01169889,
      "auxiliary_loss_mlp": 0.01029596,
      "balance_loss_clip": 0.90017104,
      "balance_loss_mlp": 1.02282202,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 1.8925364888777505,
      "language_loss": 0.72847795,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.75047284,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 2.748673439025879
    },
    {
      "auxiliary_loss_clip": 0.01164195,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 0.97310734,
      "balance_loss_mlp": 1.02344275,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 1.8905994370020898,
      "language_loss": 0.68914175,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.71109521,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.808478355407715
    },
    {
      "auxiliary_loss_clip": 0.01066578,
      "auxiliary_loss_mlp": 0.01002541,
      "balance_loss_clip": 0.89836317,
      "balance_loss_mlp": 1.00084841,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9362518796013728,
      "language_loss": 0.75345635,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77414751,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 4.27722954750061
    },
    {
      "auxiliary_loss_clip": 0.01170639,
      "auxiliary_loss_mlp": 0.01028764,
      "balance_loss_clip": 1.04851162,
      "balance_loss_mlp": 1.02138543,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 2.0434776037153766,
      "language_loss": 0.6864723,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.70846635,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.6393849849700928
    },
    {
      "auxiliary_loss_clip": 0.0116706,
      "auxiliary_loss_mlp": 0.01027278,
      "balance_loss_clip": 0.96981204,
      "balance_loss_mlp": 1.02064967,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 1.7434026240919902,
      "language_loss": 0.76633829,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.78828168,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 3.652996778488159
    },
    {
      "auxiliary_loss_clip": 0.01164688,
      "auxiliary_loss_mlp": 0.01029771,
      "balance_loss_clip": 1.00803936,
      "balance_loss_mlp": 1.02301145,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 1.7418643937352378,
      "language_loss": 0.77620941,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.798154,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.7097222805023193
    },
    {
      "auxiliary_loss_clip": 0.01165611,
      "auxiliary_loss_mlp": 0.01021189,
      "balance_loss_clip": 1.04780602,
      "balance_loss_mlp": 1.01443911,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 2.0277038883377156,
      "language_loss": 0.74902344,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.77089149,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 3.4361002445220947
    },
    {
      "auxiliary_loss_clip": 0.01166782,
      "auxiliary_loss_mlp": 0.01026137,
      "balance_loss_clip": 0.97301316,
      "balance_loss_mlp": 1.0192318,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.626042422214569,
      "language_loss": 0.72743118,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.74936038,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 2.6860318183898926
    },
    {
      "auxiliary_loss_clip": 0.01151886,
      "auxiliary_loss_mlp": 0.01121853,
      "balance_loss_clip": 0.96669066,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 2.0739346627530506,
      "language_loss": 0.68407691,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.70681429,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 2.6750733852386475
    },
    {
      "auxiliary_loss_clip": 0.01168583,
      "auxiliary_loss_mlp": 0.01028207,
      "balance_loss_clip": 0.9710201,
      "balance_loss_mlp": 1.02119732,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 2.5134743343106862,
      "language_loss": 0.6474576,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.66942555,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.614799976348877
    },
    {
      "auxiliary_loss_clip": 0.01158872,
      "auxiliary_loss_mlp": 0.01028858,
      "balance_loss_clip": 0.89122832,
      "balance_loss_mlp": 1.02184868,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 1.6971200642328266,
      "language_loss": 0.82346284,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84534013,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 2.7208242416381836
    },
    {
      "auxiliary_loss_clip": 0.01152126,
      "auxiliary_loss_mlp": 0.01022232,
      "balance_loss_clip": 0.96805364,
      "balance_loss_mlp": 1.01546383,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.5432657226163575,
      "language_loss": 0.79424232,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81598592,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 2.687452554702759
    },
    {
      "auxiliary_loss_clip": 0.01165268,
      "auxiliary_loss_mlp": 0.01122137,
      "balance_loss_clip": 0.96907759,
      "balance_loss_mlp": 0.0,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 3.1635815775002993,
      "language_loss": 0.83354592,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.85641998,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.7192165851593018
    },
    {
      "auxiliary_loss_clip": 0.01165804,
      "auxiliary_loss_mlp": 0.01028712,
      "balance_loss_clip": 1.01012373,
      "balance_loss_mlp": 1.02169919,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 1.8436969540403556,
      "language_loss": 0.67840636,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.70035154,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 2.5727622509002686
    },
    {
      "auxiliary_loss_clip": 0.0116605,
      "auxiliary_loss_mlp": 0.0102343,
      "balance_loss_clip": 1.01195168,
      "balance_loss_mlp": 1.01632822,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.4404092743613952,
      "language_loss": 0.70641959,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.7283144,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.697760581970215
    },
    {
      "auxiliary_loss_clip": 0.01164126,
      "auxiliary_loss_mlp": 0.01022692,
      "balance_loss_clip": 0.9722985,
      "balance_loss_mlp": 1.0152061,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.6765950622650794,
      "language_loss": 0.67325395,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69512218,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 2.6631407737731934
    },
    {
      "auxiliary_loss_clip": 0.011648,
      "auxiliary_loss_mlp": 0.01020601,
      "balance_loss_clip": 1.01004195,
      "balance_loss_mlp": 1.01341224,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 2.3646668681964207,
      "language_loss": 0.77280986,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.79466391,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 2.707524299621582
    },
    {
      "auxiliary_loss_clip": 0.01169537,
      "auxiliary_loss_mlp": 0.01026503,
      "balance_loss_clip": 1.04915261,
      "balance_loss_mlp": 1.01964271,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 4.035112464120941,
      "language_loss": 0.72280586,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.74476624,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 2.55295991897583
    },
    {
      "auxiliary_loss_clip": 0.01164456,
      "auxiliary_loss_mlp": 0.01020866,
      "balance_loss_clip": 0.93435729,
      "balance_loss_mlp": 1.0147028,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 1.7494840903411994,
      "language_loss": 0.70754743,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.72940058,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.7428178787231445
    },
    {
      "auxiliary_loss_clip": 0.01143652,
      "auxiliary_loss_mlp": 0.01029753,
      "balance_loss_clip": 0.85178024,
      "balance_loss_mlp": 1.02273798,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.7601806193077196,
      "language_loss": 0.77201551,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79374957,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.750436544418335
    },
    {
      "auxiliary_loss_clip": 0.0116533,
      "auxiliary_loss_mlp": 0.01024527,
      "balance_loss_clip": 0.97008538,
      "balance_loss_mlp": 1.01766026,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 1.5428126842915961,
      "language_loss": 0.72453707,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.74643564,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.625576972961426
    },
    {
      "auxiliary_loss_clip": 0.01167628,
      "auxiliary_loss_mlp": 0.01027014,
      "balance_loss_clip": 1.04838777,
      "balance_loss_mlp": 1.02006376,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.773470023073589,
      "language_loss": 0.68962866,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.71157509,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.578794240951538
    },
    {
      "auxiliary_loss_clip": 0.01156452,
      "auxiliary_loss_mlp": 0.01028155,
      "balance_loss_clip": 0.93588454,
      "balance_loss_mlp": 1.02100837,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.4248570281016506,
      "language_loss": 0.72899002,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.75083613,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.649142265319824
    },
    {
      "auxiliary_loss_clip": 0.01158585,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 0.97114158,
      "balance_loss_mlp": 1.01790154,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 1.876365105625734,
      "language_loss": 0.73180521,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75364006,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.648735761642456
    },
    {
      "auxiliary_loss_clip": 0.01165561,
      "auxiliary_loss_mlp": 0.01122853,
      "balance_loss_clip": 1.01153505,
      "balance_loss_mlp": 0.0,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 1.7215555789907115,
      "language_loss": 0.70762056,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.73050469,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 3.513004779815674
    },
    {
      "auxiliary_loss_clip": 0.01160521,
      "auxiliary_loss_mlp": 0.01021022,
      "balance_loss_clip": 0.89440441,
      "balance_loss_mlp": 1.013798,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 2.276371744241441,
      "language_loss": 0.73863471,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.76045012,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 2.687514543533325
    },
    {
      "auxiliary_loss_clip": 0.01165954,
      "auxiliary_loss_mlp": 0.01028141,
      "balance_loss_clip": 1.00904918,
      "balance_loss_mlp": 1.0216229,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 1.6239354319768464,
      "language_loss": 0.72120309,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74314404,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.6388721466064453
    },
    {
      "auxiliary_loss_clip": 0.01163113,
      "auxiliary_loss_mlp": 0.01027859,
      "balance_loss_clip": 0.89718038,
      "balance_loss_mlp": 1.02123404,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.4507869231742343,
      "language_loss": 0.77666366,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.79857337,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 2.745758533477783
    },
    {
      "auxiliary_loss_clip": 0.01169919,
      "auxiliary_loss_mlp": 0.01122691,
      "balance_loss_clip": 0.96986353,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.509815579138281,
      "language_loss": 0.77073956,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.79366571,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.6027896404266357
    },
    {
      "auxiliary_loss_clip": 0.01159536,
      "auxiliary_loss_mlp": 0.01020304,
      "balance_loss_clip": 0.89456987,
      "balance_loss_mlp": 1.01335454,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 1.5451631273686222,
      "language_loss": 0.67778552,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.69958389,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 3.752373218536377
    },
    {
      "auxiliary_loss_clip": 0.01162104,
      "auxiliary_loss_mlp": 0.01027933,
      "balance_loss_clip": 0.97061992,
      "balance_loss_mlp": 1.02088797,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.6294045008664286,
      "language_loss": 0.81785864,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.83975899,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 3.500149965286255
    },
    {
      "auxiliary_loss_clip": 0.01172923,
      "auxiliary_loss_mlp": 0.01023664,
      "balance_loss_clip": 1.01260459,
      "balance_loss_mlp": 1.01676452,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 2.236129482765967,
      "language_loss": 0.75399327,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77595913,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 3.5791831016540527
    },
    {
      "auxiliary_loss_clip": 0.01159396,
      "auxiliary_loss_mlp": 0.01025849,
      "balance_loss_clip": 0.96867824,
      "balance_loss_mlp": 1.01921153,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 1.9535340492091537,
      "language_loss": 0.73037422,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.75222665,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.758603572845459
    },
    {
      "auxiliary_loss_clip": 0.01170462,
      "auxiliary_loss_mlp": 0.01032181,
      "balance_loss_clip": 0.97294462,
      "balance_loss_mlp": 1.02478456,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 4.576745942618813,
      "language_loss": 0.71472275,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.73674917,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.7207274436950684
    },
    {
      "auxiliary_loss_clip": 0.01160873,
      "auxiliary_loss_mlp": 0.01122008,
      "balance_loss_clip": 0.9348104,
      "balance_loss_mlp": 0.0,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 2.726474420806786,
      "language_loss": 0.79765731,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.82048619,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 2.7174689769744873
    },
    {
      "auxiliary_loss_clip": 0.01162115,
      "auxiliary_loss_mlp": 0.01020998,
      "balance_loss_clip": 1.04597187,
      "balance_loss_mlp": 1.01430178,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.5491948954788566,
      "language_loss": 0.78186512,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.80369622,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.5881412029266357
    },
    {
      "auxiliary_loss_clip": 0.01164318,
      "auxiliary_loss_mlp": 0.01025869,
      "balance_loss_clip": 1.00864828,
      "balance_loss_mlp": 1.01854062,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.5130904830939187,
      "language_loss": 0.84859049,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87049234,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.7825775146484375
    },
    {
      "auxiliary_loss_clip": 0.01158797,
      "auxiliary_loss_mlp": 0.01022091,
      "balance_loss_clip": 1.00606704,
      "balance_loss_mlp": 1.01509976,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 3.093404391961825,
      "language_loss": 0.75318056,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.77498943,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 2.668793201446533
    },
    {
      "auxiliary_loss_clip": 0.01155395,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 0.85043907,
      "balance_loss_mlp": 1.01973939,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 1.889247154608036,
      "language_loss": 0.94571054,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.96753389,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.837505340576172
    },
    {
      "auxiliary_loss_clip": 0.01168312,
      "auxiliary_loss_mlp": 0.01030741,
      "balance_loss_clip": 1.0113821,
      "balance_loss_mlp": 1.02284312,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 1.654777016233181,
      "language_loss": 0.77171522,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79370576,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 2.718472480773926
    },
    {
      "auxiliary_loss_clip": 0.01166048,
      "auxiliary_loss_mlp": 0.01023535,
      "balance_loss_clip": 1.00741982,
      "balance_loss_mlp": 1.01635253,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 2.711903864361529,
      "language_loss": 0.73272467,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75462055,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.8696024417877197
    },
    {
      "auxiliary_loss_clip": 0.01062919,
      "auxiliary_loss_mlp": 0.01115798,
      "balance_loss_clip": 0.9720962,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.6733337747637349,
      "language_loss": 0.53324437,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.55503154,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 3.174959897994995
    },
    {
      "auxiliary_loss_clip": 0.01158065,
      "auxiliary_loss_mlp": 0.01026553,
      "balance_loss_clip": 0.93311334,
      "balance_loss_mlp": 1.01996088,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 1.6402525939784482,
      "language_loss": 0.73327327,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.75511944,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 2.7218358516693115
    },
    {
      "auxiliary_loss_clip": 0.01173172,
      "auxiliary_loss_mlp": 0.01025429,
      "balance_loss_clip": 1.01153767,
      "balance_loss_mlp": 1.01814508,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 3.00904216790287,
      "language_loss": 0.67384171,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.69582772,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.561352014541626
    },
    {
      "auxiliary_loss_clip": 0.01156785,
      "auxiliary_loss_mlp": 0.01020372,
      "balance_loss_clip": 0.93410301,
      "balance_loss_mlp": 1.01345158,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 1.7964071341506411,
      "language_loss": 0.78686082,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.80863237,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 2.722450017929077
    },
    {
      "auxiliary_loss_clip": 0.01169474,
      "auxiliary_loss_mlp": 0.01031943,
      "balance_loss_clip": 1.05030429,
      "balance_loss_mlp": 1.02424848,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 1.8240850266198978,
      "language_loss": 0.72098559,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74299979,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.639409303665161
    },
    {
      "auxiliary_loss_clip": 0.01163494,
      "auxiliary_loss_mlp": 0.01026971,
      "balance_loss_clip": 0.97029465,
      "balance_loss_mlp": 1.01994061,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 2.0418273255371293,
      "language_loss": 0.79366785,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.8155725,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.732257843017578
    },
    {
      "auxiliary_loss_clip": 0.01165943,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 0.97203159,
      "balance_loss_mlp": 1.01767564,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 1.9849238702550587,
      "language_loss": 0.74137485,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76327944,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.609273672103882
    },
    {
      "auxiliary_loss_clip": 0.01161965,
      "auxiliary_loss_mlp": 0.01019931,
      "balance_loss_clip": 0.93214405,
      "balance_loss_mlp": 1.01298082,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 1.8287178344053527,
      "language_loss": 0.62693989,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.64875883,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.6863179206848145
    },
    {
      "auxiliary_loss_clip": 0.01178132,
      "auxiliary_loss_mlp": 0.01027735,
      "balance_loss_clip": 1.01367164,
      "balance_loss_mlp": 1.02005208,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 1.790696230589839,
      "language_loss": 0.72820854,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.75026721,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.606003999710083
    },
    {
      "auxiliary_loss_clip": 0.01164313,
      "auxiliary_loss_mlp": 0.01028759,
      "balance_loss_clip": 1.01088142,
      "balance_loss_mlp": 1.02165425,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 2.1792676766098897,
      "language_loss": 0.80291796,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.82484871,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 3.395994186401367
    },
    {
      "auxiliary_loss_clip": 0.01166308,
      "auxiliary_loss_mlp": 0.01029085,
      "balance_loss_clip": 1.00827646,
      "balance_loss_mlp": 1.02181959,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 2.0619807020872214,
      "language_loss": 0.73378491,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.75573885,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 2.581925630569458
    },
    {
      "auxiliary_loss_clip": 0.01167194,
      "auxiliary_loss_mlp": 0.01022956,
      "balance_loss_clip": 1.00786877,
      "balance_loss_mlp": 1.01586604,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 1.9392588570694416,
      "language_loss": 0.6524784,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67437989,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.5644826889038086
    },
    {
      "auxiliary_loss_clip": 0.01161686,
      "auxiliary_loss_mlp": 0.01030072,
      "balance_loss_clip": 1.00930929,
      "balance_loss_mlp": 1.02336669,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 1.7915095828224064,
      "language_loss": 0.70888662,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.7308042,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.715796709060669
    },
    {
      "auxiliary_loss_clip": 0.01067442,
      "auxiliary_loss_mlp": 0.0100235,
      "balance_loss_clip": 0.89689022,
      "balance_loss_mlp": 1.00060964,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.8915120916881671,
      "language_loss": 0.57989049,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.60058844,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.314460039138794
    },
    {
      "auxiliary_loss_clip": 0.01152853,
      "auxiliary_loss_mlp": 0.01024932,
      "balance_loss_clip": 0.96596527,
      "balance_loss_mlp": 1.01747,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 2.6067107664490052,
      "language_loss": 0.68402255,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.70580041,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 3.52685546875
    },
    {
      "auxiliary_loss_clip": 0.01170341,
      "auxiliary_loss_mlp": 0.01032656,
      "balance_loss_clip": 1.04890037,
      "balance_loss_mlp": 1.02508032,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.7492287391728238,
      "language_loss": 0.70102644,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.72305632,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.5350630283355713
    },
    {
      "auxiliary_loss_clip": 0.01149951,
      "auxiliary_loss_mlp": 0.01021891,
      "balance_loss_clip": 0.93123496,
      "balance_loss_mlp": 1.01547146,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 1.8178149691864978,
      "language_loss": 0.79515558,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.81687397,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 3.7358922958374023
    },
    {
      "auxiliary_loss_clip": 0.0115677,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 0.92947018,
      "balance_loss_mlp": 1.01776576,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.6658717224459039,
      "language_loss": 0.70090771,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.7227236,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 3.401956796646118
    },
    {
      "auxiliary_loss_clip": 0.01156871,
      "auxiliary_loss_mlp": 0.01026541,
      "balance_loss_clip": 1.00527453,
      "balance_loss_mlp": 1.01931739,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 2.0625805469537584,
      "language_loss": 0.66461778,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68645191,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 2.6848998069763184
    },
    {
      "auxiliary_loss_clip": 0.01165269,
      "auxiliary_loss_mlp": 0.01024828,
      "balance_loss_clip": 1.00872386,
      "balance_loss_mlp": 1.01800382,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 1.789111851189495,
      "language_loss": 0.84416747,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.86606842,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 2.6429905891418457
    },
    {
      "auxiliary_loss_clip": 0.01163585,
      "auxiliary_loss_mlp": 0.01026514,
      "balance_loss_clip": 0.97453648,
      "balance_loss_mlp": 1.01961827,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 2.058212293662253,
      "language_loss": 0.86198425,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.88388526,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.634324073791504
    },
    {
      "auxiliary_loss_clip": 0.01156386,
      "auxiliary_loss_mlp": 0.01021623,
      "balance_loss_clip": 0.89257413,
      "balance_loss_mlp": 1.01490283,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 1.8562450281861493,
      "language_loss": 0.70129645,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72307652,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 2.675980567932129
    },
    {
      "auxiliary_loss_clip": 0.01155048,
      "auxiliary_loss_mlp": 0.01024141,
      "balance_loss_clip": 0.96811628,
      "balance_loss_mlp": 1.01720548,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.7434831015660825,
      "language_loss": 0.72440916,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74620104,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 2.671285390853882
    },
    {
      "auxiliary_loss_clip": 0.01164399,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.00753522,
      "balance_loss_mlp": 1.02283764,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 2.2751755235630013,
      "language_loss": 0.67944396,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.70138925,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.6061835289001465
    },
    {
      "auxiliary_loss_clip": 0.01161192,
      "auxiliary_loss_mlp": 0.01028343,
      "balance_loss_clip": 0.96721178,
      "balance_loss_mlp": 1.02167296,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 1.7422285910733941,
      "language_loss": 0.74548733,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.76738262,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 2.6946005821228027
    },
    {
      "auxiliary_loss_clip": 0.01166537,
      "auxiliary_loss_mlp": 0.01028245,
      "balance_loss_clip": 1.00748301,
      "balance_loss_mlp": 1.02099061,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 2.8902992611103113,
      "language_loss": 0.79654205,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.81848991,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 2.53898024559021
    },
    {
      "auxiliary_loss_clip": 0.01170589,
      "auxiliary_loss_mlp": 0.01027411,
      "balance_loss_clip": 1.00863659,
      "balance_loss_mlp": 1.01979923,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 1.7188051683426258,
      "language_loss": 0.70793879,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.72991872,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.758598566055298
    },
    {
      "auxiliary_loss_clip": 0.01065469,
      "auxiliary_loss_mlp": 0.01002416,
      "balance_loss_clip": 0.93390065,
      "balance_loss_mlp": 1.00072312,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7656151398965229,
      "language_loss": 0.60578299,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62646174,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.25911808013916
    },
    {
      "auxiliary_loss_clip": 0.01165776,
      "auxiliary_loss_mlp": 0.01023084,
      "balance_loss_clip": 1.04746306,
      "balance_loss_mlp": 1.0165689,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.5919384079362342,
      "language_loss": 0.73761129,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.75949991,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.579782485961914
    },
    {
      "auxiliary_loss_clip": 0.01168679,
      "auxiliary_loss_mlp": 0.01029655,
      "balance_loss_clip": 1.01258802,
      "balance_loss_mlp": 1.0228157,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 1.6950922810796432,
      "language_loss": 0.80976903,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.83175242,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.6143922805786133
    },
    {
      "auxiliary_loss_clip": 0.01169992,
      "auxiliary_loss_mlp": 0.01029211,
      "balance_loss_clip": 0.97242671,
      "balance_loss_mlp": 1.02169466,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 1.75828785133821,
      "language_loss": 0.6970427,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.71903473,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 2.8518333435058594
    },
    {
      "auxiliary_loss_clip": 0.01158723,
      "auxiliary_loss_mlp": 0.01031117,
      "balance_loss_clip": 0.93419671,
      "balance_loss_mlp": 1.02431011,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 3.7303931879527257,
      "language_loss": 0.82566857,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84756696,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.76501202583313
    },
    {
      "auxiliary_loss_clip": 0.01145043,
      "auxiliary_loss_mlp": 0.01027835,
      "balance_loss_clip": 0.92796385,
      "balance_loss_mlp": 1.02038145,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 1.3956842637280213,
      "language_loss": 0.74134803,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76307672,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.6929771900177
    },
    {
      "auxiliary_loss_clip": 0.01165683,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 0.93099844,
      "balance_loss_mlp": 1.02009058,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 2.501692195731526,
      "language_loss": 0.84435254,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86628962,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.6221582889556885
    },
    {
      "auxiliary_loss_clip": 0.01166567,
      "auxiliary_loss_mlp": 0.01022923,
      "balance_loss_clip": 0.96910179,
      "balance_loss_mlp": 1.01550567,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 1.4651681195676298,
      "language_loss": 0.69982666,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.72172153,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.759945869445801
    },
    {
      "auxiliary_loss_clip": 0.01059124,
      "auxiliary_loss_mlp": 0.01002903,
      "balance_loss_clip": 1.00826812,
      "balance_loss_mlp": 1.00129402,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7279858957601352,
      "language_loss": 0.59324276,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61386299,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 3.15122389793396
    },
    {
      "auxiliary_loss_clip": 0.01164158,
      "auxiliary_loss_mlp": 0.01122242,
      "balance_loss_clip": 0.9728564,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 1.833623979332811,
      "language_loss": 0.70824343,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.73110747,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 3.4837803840637207
    },
    {
      "auxiliary_loss_clip": 0.0117081,
      "auxiliary_loss_mlp": 0.01025143,
      "balance_loss_clip": 0.97074127,
      "balance_loss_mlp": 1.01804984,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 1.6066577552447452,
      "language_loss": 0.75620902,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77816856,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 2.6933419704437256
    },
    {
      "auxiliary_loss_clip": 0.01167734,
      "auxiliary_loss_mlp": 0.01025018,
      "balance_loss_clip": 1.00998235,
      "balance_loss_mlp": 1.01851821,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 1.8096314792879724,
      "language_loss": 0.71316898,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73509645,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.6714823246002197
    },
    {
      "auxiliary_loss_clip": 0.01168801,
      "auxiliary_loss_mlp": 0.01025324,
      "balance_loss_clip": 1.0487231,
      "balance_loss_mlp": 1.01808214,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.6699019564594932,
      "language_loss": 0.72277814,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.74471939,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.5628912448883057
    },
    {
      "auxiliary_loss_clip": 0.01162363,
      "auxiliary_loss_mlp": 0.01025058,
      "balance_loss_clip": 0.96865189,
      "balance_loss_mlp": 1.018332,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.70017568562462,
      "language_loss": 0.80107415,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82294834,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 3.5657434463500977
    },
    {
      "auxiliary_loss_clip": 0.0116412,
      "auxiliary_loss_mlp": 0.01027301,
      "balance_loss_clip": 1.00855446,
      "balance_loss_mlp": 1.02118862,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 2.10646536986411,
      "language_loss": 0.67644602,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.6983602,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 2.702023983001709
    },
    {
      "auxiliary_loss_clip": 0.01170035,
      "auxiliary_loss_mlp": 0.01026844,
      "balance_loss_clip": 0.97398138,
      "balance_loss_mlp": 1.01953697,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 1.731975891101715,
      "language_loss": 0.7326436,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.75461245,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 3.638927936553955
    },
    {
      "auxiliary_loss_clip": 0.01158775,
      "auxiliary_loss_mlp": 0.0102243,
      "balance_loss_clip": 0.96753293,
      "balance_loss_mlp": 1.01532817,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.0793404777336306,
      "language_loss": 0.6975103,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.71932232,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 3.474865198135376
    },
    {
      "auxiliary_loss_clip": 0.01145755,
      "auxiliary_loss_mlp": 0.01026073,
      "balance_loss_clip": 0.88746721,
      "balance_loss_mlp": 1.01879525,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 1.7776204614742481,
      "language_loss": 0.81731886,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.83903718,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 2.656567096710205
    },
    {
      "auxiliary_loss_clip": 0.01158406,
      "auxiliary_loss_mlp": 0.01020788,
      "balance_loss_clip": 0.93159795,
      "balance_loss_mlp": 1.01422238,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 7.507721883213795,
      "language_loss": 0.65429533,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.67608726,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 2.744656801223755
    },
    {
      "auxiliary_loss_clip": 0.01170451,
      "auxiliary_loss_mlp": 0.01025231,
      "balance_loss_clip": 1.04941988,
      "balance_loss_mlp": 1.01804304,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 1.924736733438426,
      "language_loss": 0.83304232,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.85499918,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.5641579627990723
    },
    {
      "auxiliary_loss_clip": 0.01163742,
      "auxiliary_loss_mlp": 0.01030253,
      "balance_loss_clip": 1.00878811,
      "balance_loss_mlp": 1.02244425,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 1.6124210639633143,
      "language_loss": 0.72411317,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.7460531,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 2.7026121616363525
    },
    {
      "auxiliary_loss_clip": 0.01141648,
      "auxiliary_loss_mlp": 0.01023293,
      "balance_loss_clip": 0.85327846,
      "balance_loss_mlp": 1.01625371,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 4.068800513509963,
      "language_loss": 0.74847978,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.7701292,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.9190902709960938
    },
    {
      "auxiliary_loss_clip": 0.01065803,
      "auxiliary_loss_mlp": 0.01003586,
      "balance_loss_clip": 0.89866185,
      "balance_loss_mlp": 1.00201213,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.7953684003816095,
      "language_loss": 0.58694589,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60763979,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.2619524002075195
    },
    {
      "auxiliary_loss_clip": 0.01161883,
      "auxiliary_loss_mlp": 0.01019632,
      "balance_loss_clip": 0.97379518,
      "balance_loss_mlp": 1.0129298,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.725195159473633,
      "language_loss": 0.80291986,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82473499,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 2.727447271347046
    },
    {
      "auxiliary_loss_clip": 0.01163956,
      "auxiliary_loss_mlp": 0.01028205,
      "balance_loss_clip": 1.00872397,
      "balance_loss_mlp": 1.02091861,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 2.623681125713789,
      "language_loss": 0.75282043,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.77474207,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 2.60652494430542
    },
    {
      "auxiliary_loss_clip": 0.0116331,
      "auxiliary_loss_mlp": 0.01026521,
      "balance_loss_clip": 1.04742277,
      "balance_loss_mlp": 1.0200181,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 2.176308401991881,
      "language_loss": 0.65605724,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67795557,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.548328161239624
    },
    {
      "auxiliary_loss_clip": 0.01164834,
      "auxiliary_loss_mlp": 0.01022147,
      "balance_loss_clip": 1.04446697,
      "balance_loss_mlp": 1.01474357,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.6670725193379992,
      "language_loss": 0.75407064,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.77594048,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.873410224914551
    },
    {
      "auxiliary_loss_clip": 0.0117146,
      "auxiliary_loss_mlp": 0.01030041,
      "balance_loss_clip": 1.01385927,
      "balance_loss_mlp": 1.02326357,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 1.6678783598040037,
      "language_loss": 0.79753816,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.8195532,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.64096999168396
    },
    {
      "auxiliary_loss_clip": 0.01163749,
      "auxiliary_loss_mlp": 0.01022206,
      "balance_loss_clip": 0.89524317,
      "balance_loss_mlp": 1.01543832,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 1.6840726993215118,
      "language_loss": 0.77063453,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79249418,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.6910741329193115
    },
    {
      "auxiliary_loss_clip": 0.01159329,
      "auxiliary_loss_mlp": 0.01030522,
      "balance_loss_clip": 0.93400812,
      "balance_loss_mlp": 1.02385199,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 2.2036714742175914,
      "language_loss": 0.80526233,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.82716084,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.6279609203338623
    },
    {
      "auxiliary_loss_clip": 0.01157868,
      "auxiliary_loss_mlp": 0.01024814,
      "balance_loss_clip": 0.96937108,
      "balance_loss_mlp": 1.01802242,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.6902076072632513,
      "language_loss": 0.73658609,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.75841296,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 2.815375804901123
    },
    {
      "auxiliary_loss_clip": 0.01158775,
      "auxiliary_loss_mlp": 0.01029056,
      "balance_loss_clip": 1.0080862,
      "balance_loss_mlp": 1.02190638,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 2.49792034929163,
      "language_loss": 0.82668948,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84856772,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.6243371963500977
    },
    {
      "auxiliary_loss_clip": 0.01155613,
      "auxiliary_loss_mlp": 0.01028258,
      "balance_loss_clip": 0.93179393,
      "balance_loss_mlp": 1.02105737,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 2.1672683197097298,
      "language_loss": 0.80881613,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.83065486,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.6429762840270996
    },
    {
      "auxiliary_loss_clip": 0.01171144,
      "auxiliary_loss_mlp": 0.01032962,
      "balance_loss_clip": 0.97416878,
      "balance_loss_mlp": 1.02565432,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 1.706649361586258,
      "language_loss": 0.82255089,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.84459192,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.7818572521209717
    },
    {
      "auxiliary_loss_clip": 0.01165885,
      "auxiliary_loss_mlp": 0.01027689,
      "balance_loss_clip": 1.0110333,
      "balance_loss_mlp": 1.02032232,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 1.778688589403677,
      "language_loss": 0.7924825,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.8144182,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 3.575336217880249
    },
    {
      "auxiliary_loss_clip": 0.01165094,
      "auxiliary_loss_mlp": 0.01026429,
      "balance_loss_clip": 0.89241016,
      "balance_loss_mlp": 1.01937151,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.7228140945349133,
      "language_loss": 0.73783875,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.75975394,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.730393648147583
    },
    {
      "auxiliary_loss_clip": 0.01155251,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 0.97013867,
      "balance_loss_mlp": 1.01818347,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.9044886158308798,
      "language_loss": 0.84903705,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87084162,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 2.7240231037139893
    },
    {
      "auxiliary_loss_clip": 0.01160835,
      "auxiliary_loss_mlp": 0.01026419,
      "balance_loss_clip": 1.00954568,
      "balance_loss_mlp": 1.01988673,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 1.8528624505119977,
      "language_loss": 0.78244913,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80432171,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 2.6533987522125244
    },
    {
      "auxiliary_loss_clip": 0.01163831,
      "auxiliary_loss_mlp": 0.01026064,
      "balance_loss_clip": 0.77802211,
      "balance_loss_mlp": 1.01877999,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.623145079305023,
      "language_loss": 0.60538113,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.62728,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.884411096572876
    },
    {
      "auxiliary_loss_clip": 0.0117129,
      "auxiliary_loss_mlp": 0.01028381,
      "balance_loss_clip": 0.9393003,
      "balance_loss_mlp": 1.02053428,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 2.309975847648276,
      "language_loss": 0.67764986,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.69964659,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 4.1302196979522705
    },
    {
      "auxiliary_loss_clip": 0.01164863,
      "auxiliary_loss_mlp": 0.01024714,
      "balance_loss_clip": 0.96978068,
      "balance_loss_mlp": 1.01783848,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 2.143693303101544,
      "language_loss": 0.7082119,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.73010767,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.6449122428894043
    },
    {
      "auxiliary_loss_clip": 0.01166725,
      "auxiliary_loss_mlp": 0.01026763,
      "balance_loss_clip": 0.93168956,
      "balance_loss_mlp": 1.01980674,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 1.823728614880124,
      "language_loss": 0.80475795,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82669282,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 3.715432643890381
    },
    {
      "auxiliary_loss_clip": 0.0115685,
      "auxiliary_loss_mlp": 0.01022052,
      "balance_loss_clip": 0.93272221,
      "balance_loss_mlp": 1.01526916,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 1.7252667439758278,
      "language_loss": 0.79171038,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81349939,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 3.864339828491211
    },
    {
      "auxiliary_loss_clip": 0.01155901,
      "auxiliary_loss_mlp": 0.01031141,
      "balance_loss_clip": 0.93052244,
      "balance_loss_mlp": 1.02401483,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 1.8995162466045663,
      "language_loss": 0.66405272,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.6859231,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 2.785602569580078
    },
    {
      "auxiliary_loss_clip": 0.01165866,
      "auxiliary_loss_mlp": 0.01024364,
      "balance_loss_clip": 0.9328239,
      "balance_loss_mlp": 1.01724398,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 1.9982916068711838,
      "language_loss": 0.72843945,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.75034177,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.7809700965881348
    },
    {
      "auxiliary_loss_clip": 0.01163087,
      "auxiliary_loss_mlp": 0.01019944,
      "balance_loss_clip": 0.97131127,
      "balance_loss_mlp": 1.01344681,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.7470766953570367,
      "language_loss": 0.68358409,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.70541441,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 2.6936724185943604
    },
    {
      "auxiliary_loss_clip": 0.0115564,
      "auxiliary_loss_mlp": 0.01026691,
      "balance_loss_clip": 0.89153624,
      "balance_loss_mlp": 1.02016699,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 1.7790654859122068,
      "language_loss": 0.8164677,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83829105,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 2.732235908508301
    },
    {
      "auxiliary_loss_clip": 0.011621,
      "auxiliary_loss_mlp": 0.01030003,
      "balance_loss_clip": 0.97243822,
      "balance_loss_mlp": 1.02327108,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 1.9463858961484168,
      "language_loss": 0.77469444,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79661548,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.667100667953491
    },
    {
      "auxiliary_loss_clip": 0.01163267,
      "auxiliary_loss_mlp": 0.01029934,
      "balance_loss_clip": 0.89397776,
      "balance_loss_mlp": 1.02239418,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 1.5870055691493092,
      "language_loss": 0.78658545,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.80851746,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 2.77406644821167
    },
    {
      "auxiliary_loss_clip": 0.01165696,
      "auxiliary_loss_mlp": 0.01028023,
      "balance_loss_clip": 1.04555964,
      "balance_loss_mlp": 1.02096319,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 1.7145533870476641,
      "language_loss": 0.66287661,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68481374,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 2.634188175201416
    },
    {
      "auxiliary_loss_clip": 0.01167073,
      "auxiliary_loss_mlp": 0.01028208,
      "balance_loss_clip": 1.0491761,
      "balance_loss_mlp": 1.02161241,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 1.7909072198667673,
      "language_loss": 0.79784954,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.81980234,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.6155595779418945
    },
    {
      "auxiliary_loss_clip": 0.0116372,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 0.93144459,
      "balance_loss_mlp": 1.02230823,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.0348363444356514,
      "language_loss": 0.76051402,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.78244364,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 2.77152681350708
    },
    {
      "auxiliary_loss_clip": 0.01165935,
      "auxiliary_loss_mlp": 0.01030022,
      "balance_loss_clip": 1.00758052,
      "balance_loss_mlp": 1.02300668,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 1.8435342686105798,
      "language_loss": 0.77425456,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.7962141,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.6068499088287354
    },
    {
      "auxiliary_loss_clip": 0.01151609,
      "auxiliary_loss_mlp": 0.01024801,
      "balance_loss_clip": 0.92764521,
      "balance_loss_mlp": 1.01726151,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 3.832943717198085,
      "language_loss": 0.72070754,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74247164,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.7463440895080566
    },
    {
      "auxiliary_loss_clip": 0.01164904,
      "auxiliary_loss_mlp": 0.01122271,
      "balance_loss_clip": 1.00981057,
      "balance_loss_mlp": 0.0,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 2.7948337964121968,
      "language_loss": 0.780779,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.80365074,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.6418511867523193
    },
    {
      "auxiliary_loss_clip": 0.01159818,
      "auxiliary_loss_mlp": 0.01024079,
      "balance_loss_clip": 0.92827499,
      "balance_loss_mlp": 1.01765954,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 1.6419126874510808,
      "language_loss": 0.79906023,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.82089919,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 2.6735098361968994
    },
    {
      "auxiliary_loss_clip": 0.01166262,
      "auxiliary_loss_mlp": 0.01022858,
      "balance_loss_clip": 0.96865058,
      "balance_loss_mlp": 1.01660216,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 2.034613137608446,
      "language_loss": 0.84162426,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.8635155,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.699054002761841
    },
    {
      "auxiliary_loss_clip": 0.01170831,
      "auxiliary_loss_mlp": 0.01122681,
      "balance_loss_clip": 0.97158611,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 2.3850166403346225,
      "language_loss": 0.79534394,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.81827903,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.6173980236053467
    },
    {
      "auxiliary_loss_clip": 0.01162052,
      "auxiliary_loss_mlp": 0.01024157,
      "balance_loss_clip": 0.97213811,
      "balance_loss_mlp": 1.01710594,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 2.063095095958952,
      "language_loss": 0.7042731,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72613513,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.6709537506103516
    },
    {
      "auxiliary_loss_clip": 0.01153218,
      "auxiliary_loss_mlp": 0.01122179,
      "balance_loss_clip": 1.00606036,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.1596007587626103,
      "language_loss": 0.75336945,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.77612346,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.615194797515869
    },
    {
      "auxiliary_loss_clip": 0.0116737,
      "auxiliary_loss_mlp": 0.01025802,
      "balance_loss_clip": 0.97436404,
      "balance_loss_mlp": 1.01887548,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 1.9899599962794687,
      "language_loss": 0.68419832,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70613003,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 3.5644478797912598
    },
    {
      "auxiliary_loss_clip": 0.01167139,
      "auxiliary_loss_mlp": 0.01027661,
      "balance_loss_clip": 1.00989747,
      "balance_loss_mlp": 1.02025187,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 2.304406682851743,
      "language_loss": 0.76751447,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.78946245,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.571134090423584
    },
    {
      "auxiliary_loss_clip": 0.01166564,
      "auxiliary_loss_mlp": 0.01031992,
      "balance_loss_clip": 1.04742575,
      "balance_loss_mlp": 1.02526021,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.2259352777767263,
      "language_loss": 0.65448833,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67647386,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.6826963424682617
    },
    {
      "auxiliary_loss_clip": 0.01167223,
      "auxiliary_loss_mlp": 0.01028913,
      "balance_loss_clip": 1.00744808,
      "balance_loss_mlp": 1.02179074,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.2461814900722556,
      "language_loss": 0.86110556,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.88306689,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.5911777019500732
    },
    {
      "auxiliary_loss_clip": 0.01163423,
      "auxiliary_loss_mlp": 0.01021885,
      "balance_loss_clip": 1.04829407,
      "balance_loss_mlp": 1.01569521,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.4357683340619976,
      "language_loss": 0.76464558,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78649867,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.726858139038086
    },
    {
      "auxiliary_loss_clip": 0.01167117,
      "auxiliary_loss_mlp": 0.01027561,
      "balance_loss_clip": 0.97204924,
      "balance_loss_mlp": 1.01996732,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 1.7499179460518335,
      "language_loss": 0.76793563,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.78988242,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 3.6821184158325195
    },
    {
      "auxiliary_loss_clip": 0.01157707,
      "auxiliary_loss_mlp": 0.01024896,
      "balance_loss_clip": 1.00808311,
      "balance_loss_mlp": 1.01800537,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 1.789146965405217,
      "language_loss": 0.8450824,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.86690837,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.65177845954895
    },
    {
      "auxiliary_loss_clip": 0.0116945,
      "auxiliary_loss_mlp": 0.01025035,
      "balance_loss_clip": 1.00882196,
      "balance_loss_mlp": 1.01787663,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 1.9557475051097022,
      "language_loss": 0.82644892,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.8483938,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 4.486421346664429
    },
    {
      "auxiliary_loss_clip": 0.01162617,
      "auxiliary_loss_mlp": 0.01024841,
      "balance_loss_clip": 1.00813496,
      "balance_loss_mlp": 1.01737225,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 1.8182098950769345,
      "language_loss": 0.77764481,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79951936,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 2.6022911071777344
    },
    {
      "auxiliary_loss_clip": 0.01155733,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 0.97035277,
      "balance_loss_mlp": 1.02345312,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 1.8969277798138913,
      "language_loss": 0.67246091,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69432682,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.647847890853882
    },
    {
      "auxiliary_loss_clip": 0.01157646,
      "auxiliary_loss_mlp": 0.01028196,
      "balance_loss_clip": 0.93224835,
      "balance_loss_mlp": 1.02185154,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 1.5900100898254461,
      "language_loss": 0.7925657,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.8144241,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 2.704955816268921
    },
    {
      "auxiliary_loss_clip": 0.01169805,
      "auxiliary_loss_mlp": 0.01022787,
      "balance_loss_clip": 0.97499901,
      "balance_loss_mlp": 1.01615918,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 1.8841945301202878,
      "language_loss": 0.75265419,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77458012,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 2.690849781036377
    },
    {
      "auxiliary_loss_clip": 0.01087711,
      "auxiliary_loss_mlp": 0.01116621,
      "balance_loss_clip": 0.76170421,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 1.3163899575221363,
      "language_loss": 0.64626813,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66831142,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 3.8248817920684814
    },
    {
      "auxiliary_loss_clip": 0.01168574,
      "auxiliary_loss_mlp": 0.01023396,
      "balance_loss_clip": 1.00829315,
      "balance_loss_mlp": 1.01662517,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 1.9448554162534206,
      "language_loss": 0.77536988,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79728955,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 3.7550644874572754
    },
    {
      "auxiliary_loss_clip": 0.01161922,
      "auxiliary_loss_mlp": 0.01024281,
      "balance_loss_clip": 0.97037685,
      "balance_loss_mlp": 1.01722407,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 2.0026847011555717,
      "language_loss": 0.81026316,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.83212519,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.6786344051361084
    },
    {
      "auxiliary_loss_clip": 0.01159938,
      "auxiliary_loss_mlp": 0.01024641,
      "balance_loss_clip": 0.93268281,
      "balance_loss_mlp": 1.01791453,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 1.960032839571581,
      "language_loss": 0.76855612,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79040194,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 2.722126007080078
    },
    {
      "auxiliary_loss_clip": 0.01155891,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 0.89390302,
      "balance_loss_mlp": 1.01882601,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 2.0232520663996674,
      "language_loss": 0.78059196,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80241656,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.732550621032715
    },
    {
      "auxiliary_loss_clip": 0.01167862,
      "auxiliary_loss_mlp": 0.0112307,
      "balance_loss_clip": 1.01240611,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 1.7660428862752542,
      "language_loss": 0.7625584,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.78546774,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 2.6898319721221924
    },
    {
      "auxiliary_loss_clip": 0.01160587,
      "auxiliary_loss_mlp": 0.01022387,
      "balance_loss_clip": 1.00811684,
      "balance_loss_mlp": 1.01524031,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 1.7079350057601457,
      "language_loss": 0.766271,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.78810072,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.582920789718628
    },
    {
      "auxiliary_loss_clip": 0.01168181,
      "auxiliary_loss_mlp": 0.01022431,
      "balance_loss_clip": 1.04664457,
      "balance_loss_mlp": 1.01540399,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 2.0748478983814467,
      "language_loss": 0.77964795,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.80155408,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.625716209411621
    },
    {
      "auxiliary_loss_clip": 0.01158589,
      "auxiliary_loss_mlp": 0.01122713,
      "balance_loss_clip": 0.92997491,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 1.9955884085953834,
      "language_loss": 0.79583514,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.81864816,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.627789258956909
    },
    {
      "auxiliary_loss_clip": 0.01164278,
      "auxiliary_loss_mlp": 0.01019988,
      "balance_loss_clip": 0.96786773,
      "balance_loss_mlp": 1.0132854,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 2.246904191169017,
      "language_loss": 0.83152533,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85336804,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.647242546081543
    },
    {
      "auxiliary_loss_clip": 0.01169799,
      "auxiliary_loss_mlp": 0.01026954,
      "balance_loss_clip": 1.01216817,
      "balance_loss_mlp": 1.01982796,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 2.668107289026472,
      "language_loss": 0.76161695,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.78358448,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 2.600123405456543
    },
    {
      "auxiliary_loss_clip": 0.01164754,
      "auxiliary_loss_mlp": 0.0102729,
      "balance_loss_clip": 0.97272384,
      "balance_loss_mlp": 1.01986337,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 1.826671110580632,
      "language_loss": 0.65816164,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.68008208,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 2.9035143852233887
    },
    {
      "auxiliary_loss_clip": 0.01161074,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.00734186,
      "balance_loss_mlp": 1.02043557,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 1.863715694642821,
      "language_loss": 0.75049365,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.7723785,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.7038023471832275
    },
    {
      "auxiliary_loss_clip": 0.0116632,
      "auxiliary_loss_mlp": 0.01029529,
      "balance_loss_clip": 1.00920033,
      "balance_loss_mlp": 1.0226984,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 1.5247611127679632,
      "language_loss": 0.75511295,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.77707142,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.649099111557007
    },
    {
      "auxiliary_loss_clip": 0.01058987,
      "auxiliary_loss_mlp": 0.01001126,
      "balance_loss_clip": 1.00806665,
      "balance_loss_mlp": 0.99951667,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 1.1653711123536454,
      "language_loss": 0.57589489,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59649599,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 3.9082627296447754
    },
    {
      "auxiliary_loss_clip": 0.01163191,
      "auxiliary_loss_mlp": 0.0102473,
      "balance_loss_clip": 1.00725925,
      "balance_loss_mlp": 1.01779211,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 2.0096475158593963,
      "language_loss": 0.72886634,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.75074553,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.625314950942993
    },
    {
      "auxiliary_loss_clip": 0.01161177,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 0.85530102,
      "balance_loss_mlp": 1.02287269,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 2.821136361594868,
      "language_loss": 0.58501297,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60692841,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.783538579940796
    },
    {
      "auxiliary_loss_clip": 0.01069785,
      "auxiliary_loss_mlp": 0.0100384,
      "balance_loss_clip": 0.86045825,
      "balance_loss_mlp": 1.00214756,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7399106166013997,
      "language_loss": 0.55166173,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57239795,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.384188652038574
    },
    {
      "auxiliary_loss_clip": 0.01165047,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 0.97041011,
      "balance_loss_mlp": 1.01883388,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.9975979245883533,
      "language_loss": 0.69789708,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.71980214,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 3.2944581508636475
    },
    {
      "auxiliary_loss_clip": 0.01171372,
      "auxiliary_loss_mlp": 0.01027918,
      "balance_loss_clip": 1.04882264,
      "balance_loss_mlp": 1.01989532,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.8938260306500339,
      "language_loss": 0.71356416,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.73555702,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 3.6956827640533447
    },
    {
      "auxiliary_loss_clip": 0.01170305,
      "auxiliary_loss_mlp": 0.01032661,
      "balance_loss_clip": 1.04990101,
      "balance_loss_mlp": 1.02480483,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 2.6628230531566075,
      "language_loss": 0.68730295,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.70933259,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.5650038719177246
    },
    {
      "auxiliary_loss_clip": 0.01164597,
      "auxiliary_loss_mlp": 0.01024473,
      "balance_loss_clip": 1.01041436,
      "balance_loss_mlp": 1.01726353,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 2.3848419456608676,
      "language_loss": 0.79549122,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.81738198,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 3.588890314102173
    },
    {
      "auxiliary_loss_clip": 0.01152031,
      "auxiliary_loss_mlp": 0.01025177,
      "balance_loss_clip": 0.96854329,
      "balance_loss_mlp": 1.01854873,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 1.6910617628996287,
      "language_loss": 0.71553493,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.73730707,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 2.6488430500030518
    },
    {
      "auxiliary_loss_clip": 0.01165611,
      "auxiliary_loss_mlp": 0.01028685,
      "balance_loss_clip": 1.00939035,
      "balance_loss_mlp": 1.02173519,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 1.8561791957614961,
      "language_loss": 0.63453376,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65647674,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 2.5671615600585938
    },
    {
      "auxiliary_loss_clip": 0.01168415,
      "auxiliary_loss_mlp": 0.01027975,
      "balance_loss_clip": 1.00931895,
      "balance_loss_mlp": 1.02117157,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.001164568300307,
      "language_loss": 0.69299912,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71496308,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.5965890884399414
    },
    {
      "auxiliary_loss_clip": 0.01166344,
      "auxiliary_loss_mlp": 0.01025203,
      "balance_loss_clip": 1.04728508,
      "balance_loss_mlp": 1.01788068,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 1.820490698709031,
      "language_loss": 0.7349714,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.7568869,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.565706253051758
    },
    {
      "auxiliary_loss_clip": 0.01164664,
      "auxiliary_loss_mlp": 0.01020816,
      "balance_loss_clip": 0.97185409,
      "balance_loss_mlp": 1.01399159,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 2.529657002636531,
      "language_loss": 0.78655815,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.80841297,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 2.6340081691741943
    },
    {
      "auxiliary_loss_clip": 0.01167379,
      "auxiliary_loss_mlp": 0.01025502,
      "balance_loss_clip": 1.01042163,
      "balance_loss_mlp": 1.01892471,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 2.5752020958593262,
      "language_loss": 0.85844707,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88037586,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 2.63224458694458
    },
    {
      "auxiliary_loss_clip": 0.01166719,
      "auxiliary_loss_mlp": 0.01023035,
      "balance_loss_clip": 1.00822592,
      "balance_loss_mlp": 1.01584053,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.500179014869186,
      "language_loss": 0.75653815,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.77843571,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.595547676086426
    },
    {
      "auxiliary_loss_clip": 0.01070226,
      "auxiliary_loss_mlp": 0.01003113,
      "balance_loss_clip": 0.89659792,
      "balance_loss_mlp": 1.00134873,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7562748958238621,
      "language_loss": 0.63060796,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.65134132,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 3.2474186420440674
    },
    {
      "auxiliary_loss_clip": 0.01155913,
      "auxiliary_loss_mlp": 0.01024676,
      "balance_loss_clip": 0.89178276,
      "balance_loss_mlp": 1.01771462,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.06996813400915,
      "language_loss": 0.81479996,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.83660585,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 2.7241928577423096
    },
    {
      "auxiliary_loss_clip": 0.0116102,
      "auxiliary_loss_mlp": 0.01023628,
      "balance_loss_clip": 0.9310801,
      "balance_loss_mlp": 1.01692259,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.0821353674443883,
      "language_loss": 0.87726545,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.89911193,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 2.6898579597473145
    },
    {
      "auxiliary_loss_clip": 0.01169657,
      "auxiliary_loss_mlp": 0.01022805,
      "balance_loss_clip": 1.01067519,
      "balance_loss_mlp": 1.01554513,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 2.3275012576717846,
      "language_loss": 0.55630028,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.5782249,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.694977045059204
    },
    {
      "auxiliary_loss_clip": 0.01168763,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.01117229,
      "balance_loss_mlp": 1.01859188,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 2.0830575414530843,
      "language_loss": 0.77365339,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.79559499,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.6515467166900635
    },
    {
      "auxiliary_loss_clip": 0.01168083,
      "auxiliary_loss_mlp": 0.01022319,
      "balance_loss_clip": 0.97074342,
      "balance_loss_mlp": 1.01573586,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.5753014009729063,
      "language_loss": 0.81108093,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.83298492,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.709538698196411
    },
    {
      "auxiliary_loss_clip": 0.01164906,
      "auxiliary_loss_mlp": 0.01024373,
      "balance_loss_clip": 0.9740051,
      "balance_loss_mlp": 1.01753008,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 2.1120231463242884,
      "language_loss": 0.69665778,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.71855056,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.64255428314209
    },
    {
      "auxiliary_loss_clip": 0.01168449,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.01054084,
      "balance_loss_mlp": 1.02026105,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.3486918472418266,
      "language_loss": 0.66866237,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.69062614,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.651668071746826
    },
    {
      "auxiliary_loss_clip": 0.01167439,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.0086143,
      "balance_loss_mlp": 1.01889408,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 2.41969467533508,
      "language_loss": 0.82812208,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.85006011,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 2.682593822479248
    },
    {
      "auxiliary_loss_clip": 0.01164744,
      "auxiliary_loss_mlp": 0.01122652,
      "balance_loss_clip": 0.96945173,
      "balance_loss_mlp": 0.0,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 1.9355281184157715,
      "language_loss": 0.7080766,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.73095059,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.6555428504943848
    },
    {
      "auxiliary_loss_clip": 0.01165258,
      "auxiliary_loss_mlp": 0.01122805,
      "balance_loss_clip": 0.93204343,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 2.423931862247547,
      "language_loss": 0.70735955,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.73024017,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.7725181579589844
    },
    {
      "auxiliary_loss_clip": 0.01167647,
      "auxiliary_loss_mlp": 0.01021874,
      "balance_loss_clip": 0.93486297,
      "balance_loss_mlp": 1.01524019,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.652101958973146,
      "language_loss": 0.69434106,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71623629,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 3.5820751190185547
    },
    {
      "auxiliary_loss_clip": 0.01163823,
      "auxiliary_loss_mlp": 0.01030501,
      "balance_loss_clip": 0.96984076,
      "balance_loss_mlp": 1.02359569,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.7575983354803457,
      "language_loss": 0.74237704,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76432031,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.7119503021240234
    },
    {
      "auxiliary_loss_clip": 0.01056033,
      "auxiliary_loss_mlp": 0.01115544,
      "balance_loss_clip": 0.96959394,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7252974599942166,
      "language_loss": 0.57578504,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.5975008,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.2311389446258545
    },
    {
      "auxiliary_loss_clip": 0.01160424,
      "auxiliary_loss_mlp": 0.01026312,
      "balance_loss_clip": 0.96949196,
      "balance_loss_mlp": 1.01911163,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 3.397318434954599,
      "language_loss": 0.79159898,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81346631,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.633563995361328
    },
    {
      "auxiliary_loss_clip": 0.01162073,
      "auxiliary_loss_mlp": 0.01024072,
      "balance_loss_clip": 1.00962067,
      "balance_loss_mlp": 1.01744652,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.5319784078917658,
      "language_loss": 0.76884407,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.7907055,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.6393370628356934
    },
    {
      "auxiliary_loss_clip": 0.01167489,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 0.85879278,
      "balance_loss_mlp": 1.01748633,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 1.7761743254624487,
      "language_loss": 0.88031888,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90224582,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.702803373336792
    },
    {
      "auxiliary_loss_clip": 0.01169984,
      "auxiliary_loss_mlp": 0.01030522,
      "balance_loss_clip": 1.01023591,
      "balance_loss_mlp": 1.02273798,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 2.338066473260236,
      "language_loss": 0.72464299,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.74664807,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 3.6398181915283203
    },
    {
      "auxiliary_loss_clip": 0.011682,
      "auxiliary_loss_mlp": 0.01023684,
      "balance_loss_clip": 1.04897261,
      "balance_loss_mlp": 1.01675224,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.0384272726484323,
      "language_loss": 0.72577918,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.74769807,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 3.555964708328247
    },
    {
      "auxiliary_loss_clip": 0.01168612,
      "auxiliary_loss_mlp": 0.01122147,
      "balance_loss_clip": 0.97002518,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 4.001097892398639,
      "language_loss": 0.70987743,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.73278499,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 3.5781068801879883
    },
    {
      "auxiliary_loss_clip": 0.01162885,
      "auxiliary_loss_mlp": 0.01023855,
      "balance_loss_clip": 0.97173512,
      "balance_loss_mlp": 1.01700628,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 1.6959510104822793,
      "language_loss": 0.80961931,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.8314867,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 2.6521058082580566
    },
    {
      "auxiliary_loss_clip": 0.01170915,
      "auxiliary_loss_mlp": 0.0103253,
      "balance_loss_clip": 1.04913557,
      "balance_loss_mlp": 1.0252049,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.123695322616834,
      "language_loss": 0.67556739,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.69760191,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.5380892753601074
    },
    {
      "auxiliary_loss_clip": 0.01146082,
      "auxiliary_loss_mlp": 0.01026542,
      "balance_loss_clip": 0.92848754,
      "balance_loss_mlp": 1.01964283,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 1.7641686247584163,
      "language_loss": 0.71705198,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.73877823,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 2.668782949447632
    },
    {
      "auxiliary_loss_clip": 0.01161299,
      "auxiliary_loss_mlp": 0.01024902,
      "balance_loss_clip": 0.93108189,
      "balance_loss_mlp": 1.01849127,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 1.6768820963766486,
      "language_loss": 0.74287951,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76474148,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.660231351852417
    },
    {
      "auxiliary_loss_clip": 0.0115737,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 0.96982682,
      "balance_loss_mlp": 1.01821017,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 1.7560843842224938,
      "language_loss": 0.81490445,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.83673352,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 2.6098849773406982
    },
    {
      "auxiliary_loss_clip": 0.01157935,
      "auxiliary_loss_mlp": 0.010235,
      "balance_loss_clip": 0.93304598,
      "balance_loss_mlp": 1.01689327,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 2.0815934151197424,
      "language_loss": 0.76689893,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78871334,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.7566468715667725
    },
    {
      "auxiliary_loss_clip": 0.0117074,
      "auxiliary_loss_mlp": 0.01026551,
      "balance_loss_clip": 0.97194642,
      "balance_loss_mlp": 1.01959813,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 1.6465974999627633,
      "language_loss": 0.74539065,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76736355,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 2.6716177463531494
    },
    {
      "auxiliary_loss_clip": 0.01162644,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.00964725,
      "balance_loss_mlp": 1.01745129,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.2012316193881505,
      "language_loss": 0.72142386,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74329233,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.6271321773529053
    },
    {
      "auxiliary_loss_clip": 0.01157092,
      "auxiliary_loss_mlp": 0.01121619,
      "balance_loss_clip": 0.96875048,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 2.2061938443053988,
      "language_loss": 0.79954493,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.82233202,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 2.6311581134796143
    },
    {
      "auxiliary_loss_clip": 0.01165956,
      "auxiliary_loss_mlp": 0.01025228,
      "balance_loss_clip": 1.04883516,
      "balance_loss_mlp": 1.01822138,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.6677171587834276,
      "language_loss": 0.68061376,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70252556,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.590041399002075
    },
    {
      "auxiliary_loss_clip": 0.01170698,
      "auxiliary_loss_mlp": 0.01027354,
      "balance_loss_clip": 1.04836774,
      "balance_loss_mlp": 1.02015948,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 4.33372974063976,
      "language_loss": 0.76567817,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.78765875,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.6033947467803955
    },
    {
      "auxiliary_loss_clip": 0.0115349,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 0.96973264,
      "balance_loss_mlp": 1.01818871,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 3.18306455270663,
      "language_loss": 0.70061505,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.72240549,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.850520133972168
    },
    {
      "auxiliary_loss_clip": 0.01164925,
      "auxiliary_loss_mlp": 0.01023596,
      "balance_loss_clip": 1.0120014,
      "balance_loss_mlp": 1.01653838,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.5872827229059494,
      "language_loss": 0.81714368,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.83902884,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.619040012359619
    },
    {
      "auxiliary_loss_clip": 0.01173186,
      "auxiliary_loss_mlp": 0.01122885,
      "balance_loss_clip": 1.01151645,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 1.7398057871908506,
      "language_loss": 0.78802013,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.8109808,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.703259229660034
    },
    {
      "auxiliary_loss_clip": 0.01164582,
      "auxiliary_loss_mlp": 0.01023625,
      "balance_loss_clip": 0.85658473,
      "balance_loss_mlp": 1.0169909,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 1.9353221366620135,
      "language_loss": 0.70699888,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.72888094,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 2.9258010387420654
    },
    {
      "auxiliary_loss_clip": 0.01168991,
      "auxiliary_loss_mlp": 0.01025645,
      "balance_loss_clip": 1.00745595,
      "balance_loss_mlp": 1.018507,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 1.949325611572555,
      "language_loss": 0.62995452,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.65190089,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 2.6395578384399414
    },
    {
      "auxiliary_loss_clip": 0.0116203,
      "auxiliary_loss_mlp": 0.01024082,
      "balance_loss_clip": 0.92894596,
      "balance_loss_mlp": 1.01670313,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 2.036704685921344,
      "language_loss": 0.76399326,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.7858544,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.703270673751831
    },
    {
      "auxiliary_loss_clip": 0.01170134,
      "auxiliary_loss_mlp": 0.01021657,
      "balance_loss_clip": 1.01043594,
      "balance_loss_mlp": 1.01471353,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 1.8707674591241659,
      "language_loss": 0.79297727,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81489515,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 3.376819610595703
    },
    {
      "auxiliary_loss_clip": 0.01155338,
      "auxiliary_loss_mlp": 0.01021022,
      "balance_loss_clip": 0.92936337,
      "balance_loss_mlp": 1.01438475,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 1.7567953044333453,
      "language_loss": 0.78067267,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.80243635,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.726440906524658
    },
    {
      "auxiliary_loss_clip": 0.01170457,
      "auxiliary_loss_mlp": 0.01122508,
      "balance_loss_clip": 0.93085539,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 2.194547462199194,
      "language_loss": 0.75094128,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.77387094,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.647730588912964
    },
    {
      "auxiliary_loss_clip": 0.01163447,
      "auxiliary_loss_mlp": 0.0102456,
      "balance_loss_clip": 1.0080719,
      "balance_loss_mlp": 1.01822066,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 2.8439823762672405,
      "language_loss": 0.73582262,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.75770265,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.611541748046875
    },
    {
      "auxiliary_loss_clip": 0.01158893,
      "auxiliary_loss_mlp": 0.01020772,
      "balance_loss_clip": 0.93456316,
      "balance_loss_mlp": 1.01407886,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 1.8208767144728784,
      "language_loss": 0.64384997,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66564667,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.6822924613952637
    },
    {
      "auxiliary_loss_clip": 0.01157289,
      "auxiliary_loss_mlp": 0.01026901,
      "balance_loss_clip": 0.9708457,
      "balance_loss_mlp": 1.01949239,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 1.7771939197581768,
      "language_loss": 0.88461411,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90645605,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 2.7424683570861816
    },
    {
      "auxiliary_loss_clip": 0.01163729,
      "auxiliary_loss_mlp": 0.01024726,
      "balance_loss_clip": 1.00997996,
      "balance_loss_mlp": 1.018116,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 1.6693228667563504,
      "language_loss": 0.66971719,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.69160175,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 3.6298739910125732
    },
    {
      "auxiliary_loss_clip": 0.01167069,
      "auxiliary_loss_mlp": 0.01121864,
      "balance_loss_clip": 1.01026392,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 1.7133646488466778,
      "language_loss": 0.8137604,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.83664972,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 3.611258029937744
    },
    {
      "auxiliary_loss_clip": 0.01164074,
      "auxiliary_loss_mlp": 0.01027278,
      "balance_loss_clip": 1.00830019,
      "balance_loss_mlp": 1.01984203,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 1.9183803028256943,
      "language_loss": 0.80113935,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.82305288,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 2.700687885284424
    },
    {
      "auxiliary_loss_clip": 0.01161104,
      "auxiliary_loss_mlp": 0.01023818,
      "balance_loss_clip": 0.97127008,
      "balance_loss_mlp": 1.01732719,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 1.9953564747403476,
      "language_loss": 0.70123881,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.72308803,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.6479907035827637
    },
    {
      "auxiliary_loss_clip": 0.01058863,
      "auxiliary_loss_mlp": 0.01001724,
      "balance_loss_clip": 1.00801873,
      "balance_loss_mlp": 1.00012636,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8199544908270884,
      "language_loss": 0.60590541,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62651128,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 3.332338571548462
    },
    {
      "auxiliary_loss_clip": 0.01171168,
      "auxiliary_loss_mlp": 0.01024063,
      "balance_loss_clip": 0.97204876,
      "balance_loss_mlp": 1.01662493,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.773031791778122,
      "language_loss": 0.71550667,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.73745894,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 2.624748706817627
    },
    {
      "auxiliary_loss_clip": 0.01165895,
      "auxiliary_loss_mlp": 0.01024908,
      "balance_loss_clip": 1.04857695,
      "balance_loss_mlp": 1.01760972,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.668357414422252,
      "language_loss": 0.62629306,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64820105,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.5712647438049316
    },
    {
      "auxiliary_loss_clip": 0.01163279,
      "auxiliary_loss_mlp": 0.01021111,
      "balance_loss_clip": 0.89523661,
      "balance_loss_mlp": 1.01427436,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 1.9189821525493267,
      "language_loss": 0.85053629,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87238014,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.65090274810791
    },
    {
      "auxiliary_loss_clip": 0.01164685,
      "auxiliary_loss_mlp": 0.01023981,
      "balance_loss_clip": 1.0080657,
      "balance_loss_mlp": 1.01703691,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 1.5172792170211369,
      "language_loss": 0.68287015,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.70475686,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 2.624241352081299
    },
    {
      "auxiliary_loss_clip": 0.01159998,
      "auxiliary_loss_mlp": 0.0103176,
      "balance_loss_clip": 0.97160202,
      "balance_loss_mlp": 1.0248611,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.5584562058634455,
      "language_loss": 0.81006622,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.8319838,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 2.6616432666778564
    },
    {
      "auxiliary_loss_clip": 0.01157065,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.00708818,
      "balance_loss_mlp": 1.02071309,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.232264875925813,
      "language_loss": 0.75291371,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77476513,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 2.6309733390808105
    },
    {
      "auxiliary_loss_clip": 0.01158617,
      "auxiliary_loss_mlp": 0.01027531,
      "balance_loss_clip": 0.96874475,
      "balance_loss_mlp": 1.02054489,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 2.280949873580767,
      "language_loss": 0.70319307,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.7250545,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 2.6720597743988037
    },
    {
      "auxiliary_loss_clip": 0.01169166,
      "auxiliary_loss_mlp": 0.01025262,
      "balance_loss_clip": 0.93301487,
      "balance_loss_mlp": 1.01798689,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.5887695166524742,
      "language_loss": 0.73349178,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75543606,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.7048614025115967
    },
    {
      "auxiliary_loss_clip": 0.01165394,
      "auxiliary_loss_mlp": 0.0103095,
      "balance_loss_clip": 1.04890645,
      "balance_loss_mlp": 1.02376461,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 1.9602267769544317,
      "language_loss": 0.83367509,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.8556385,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.6767466068267822
    },
    {
      "auxiliary_loss_clip": 0.01165187,
      "auxiliary_loss_mlp": 0.01029527,
      "balance_loss_clip": 1.00893044,
      "balance_loss_mlp": 1.02268195,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 2.0537786841039267,
      "language_loss": 0.66338253,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.68532968,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.5898971557617188
    },
    {
      "auxiliary_loss_clip": 0.01169404,
      "auxiliary_loss_mlp": 0.01025347,
      "balance_loss_clip": 1.04714656,
      "balance_loss_mlp": 1.01736307,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 1.9673140561544122,
      "language_loss": 0.79014027,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.81208777,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.663294792175293
    },
    {
      "auxiliary_loss_clip": 0.01159346,
      "auxiliary_loss_mlp": 0.01026619,
      "balance_loss_clip": 0.93140519,
      "balance_loss_mlp": 1.01993692,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 1.5611058120466916,
      "language_loss": 0.74182469,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76368427,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.732111930847168
    },
    {
      "auxiliary_loss_clip": 0.01159848,
      "auxiliary_loss_mlp": 0.01029607,
      "balance_loss_clip": 1.00629938,
      "balance_loss_mlp": 1.02212667,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 1.8099759996327351,
      "language_loss": 0.71364397,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.73553848,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 2.701341152191162
    },
    {
      "auxiliary_loss_clip": 0.0107503,
      "auxiliary_loss_mlp": 0.01001533,
      "balance_loss_clip": 0.947074,
      "balance_loss_mlp": 0.99982816,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.9112032228756429,
      "language_loss": 0.61257178,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63333738,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.296916961669922
    },
    {
      "auxiliary_loss_clip": 0.0116579,
      "auxiliary_loss_mlp": 0.01122331,
      "balance_loss_clip": 0.93717909,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.881447944531793,
      "language_loss": 0.72616142,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.74904263,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 3.0766537189483643
    },
    {
      "auxiliary_loss_clip": 0.01167072,
      "auxiliary_loss_mlp": 0.01024816,
      "balance_loss_clip": 0.93407011,
      "balance_loss_mlp": 1.01808083,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 3.5514274635127334,
      "language_loss": 0.70203257,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72395146,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 3.442328453063965
    },
    {
      "auxiliary_loss_clip": 0.01165303,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.04791999,
      "balance_loss_mlp": 1.01817226,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 2.019899973858902,
      "language_loss": 0.77857828,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.80047894,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.6336190700531006
    },
    {
      "auxiliary_loss_clip": 0.01164648,
      "auxiliary_loss_mlp": 0.01019376,
      "balance_loss_clip": 1.04692149,
      "balance_loss_mlp": 1.01289964,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.815828183285261,
      "language_loss": 0.84501988,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86686009,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.525911331176758
    },
    {
      "auxiliary_loss_clip": 0.01142502,
      "auxiliary_loss_mlp": 0.01029033,
      "balance_loss_clip": 0.88998616,
      "balance_loss_mlp": 1.02253616,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 2.0043982672480523,
      "language_loss": 0.71174037,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.73345578,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.781878709793091
    },
    {
      "auxiliary_loss_clip": 0.01170094,
      "auxiliary_loss_mlp": 0.01022644,
      "balance_loss_clip": 1.00916767,
      "balance_loss_mlp": 1.01539052,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 2.6277426381388613,
      "language_loss": 0.65635002,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.67827737,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.666161298751831
    },
    {
      "auxiliary_loss_clip": 0.01169067,
      "auxiliary_loss_mlp": 0.01024503,
      "balance_loss_clip": 0.89445704,
      "balance_loss_mlp": 1.01763654,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 1.9114890728692453,
      "language_loss": 0.82411194,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84604758,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 3.703435182571411
    },
    {
      "auxiliary_loss_clip": 0.01162018,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 0.97055483,
      "balance_loss_mlp": 1.01747549,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 1.7418715238187092,
      "language_loss": 0.79493582,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.81680411,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 2.6057894229888916
    },
    {
      "auxiliary_loss_clip": 0.01060581,
      "auxiliary_loss_mlp": 0.01115574,
      "balance_loss_clip": 0.93577933,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.7961973243150018,
      "language_loss": 0.56482828,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58658981,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 4.27347469329834
    },
    {
      "auxiliary_loss_clip": 0.01167852,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 1.01370347,
      "balance_loss_mlp": 1.02047682,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 1.647637558391886,
      "language_loss": 0.76347136,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78542399,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 2.7043066024780273
    },
    {
      "auxiliary_loss_clip": 0.01167291,
      "auxiliary_loss_mlp": 0.01022736,
      "balance_loss_clip": 1.04856062,
      "balance_loss_mlp": 1.01610184,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 1.7526667919242067,
      "language_loss": 0.85316539,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87506568,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 2.5512163639068604
    },
    {
      "auxiliary_loss_clip": 0.01168243,
      "auxiliary_loss_mlp": 0.01025685,
      "balance_loss_clip": 1.00895154,
      "balance_loss_mlp": 1.01873553,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 1.9708791151602356,
      "language_loss": 0.64521813,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.66715747,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.6922500133514404
    },
    {
      "auxiliary_loss_clip": 0.01170129,
      "auxiliary_loss_mlp": 0.01022861,
      "balance_loss_clip": 1.05092371,
      "balance_loss_mlp": 1.0158608,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.5933086140787958,
      "language_loss": 0.73703593,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75896585,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 2.5240390300750732
    },
    {
      "auxiliary_loss_clip": 0.01159976,
      "auxiliary_loss_mlp": 0.010228,
      "balance_loss_clip": 0.97187436,
      "balance_loss_mlp": 1.01629996,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.8265322039737169,
      "language_loss": 0.7758401,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79766786,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 2.754866361618042
    },
    {
      "auxiliary_loss_clip": 0.01166235,
      "auxiliary_loss_mlp": 0.01026701,
      "balance_loss_clip": 1.00941491,
      "balance_loss_mlp": 1.02009678,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 5.810471405494502,
      "language_loss": 0.83220422,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85413361,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.5540857315063477
    },
    {
      "auxiliary_loss_clip": 0.01160884,
      "auxiliary_loss_mlp": 0.01033103,
      "balance_loss_clip": 1.00976789,
      "balance_loss_mlp": 1.02587295,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 2.0982008541511736,
      "language_loss": 0.63694358,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.65888345,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.5995993614196777
    },
    {
      "auxiliary_loss_clip": 0.0116259,
      "auxiliary_loss_mlp": 0.01029708,
      "balance_loss_clip": 0.93085921,
      "balance_loss_mlp": 1.02253795,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 1.497984569124537,
      "language_loss": 0.72731316,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.74923611,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 2.6880786418914795
    },
    {
      "auxiliary_loss_clip": 0.01168492,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.01014924,
      "balance_loss_mlp": 1.02170944,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 2.3760238292061353,
      "language_loss": 0.75742137,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.77939475,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 2.585024118423462
    },
    {
      "auxiliary_loss_clip": 0.01163036,
      "auxiliary_loss_mlp": 0.01024019,
      "balance_loss_clip": 1.00815415,
      "balance_loss_mlp": 1.01744127,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 1.766855521955724,
      "language_loss": 0.65176964,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.67364019,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 2.587415933609009
    },
    {
      "auxiliary_loss_clip": 0.01165156,
      "auxiliary_loss_mlp": 0.01026299,
      "balance_loss_clip": 0.89815116,
      "balance_loss_mlp": 1.01951897,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.418089017629428,
      "language_loss": 0.79220819,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.8141228,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.6641552448272705
    },
    {
      "auxiliary_loss_clip": 0.01159737,
      "auxiliary_loss_mlp": 0.01030398,
      "balance_loss_clip": 1.00922132,
      "balance_loss_mlp": 1.02323961,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.846987759988745,
      "language_loss": 0.77361631,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79551768,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.6986286640167236
    },
    {
      "auxiliary_loss_clip": 0.01171609,
      "auxiliary_loss_mlp": 0.01022261,
      "balance_loss_clip": 1.01017356,
      "balance_loss_mlp": 1.01472151,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 2.416207049935834,
      "language_loss": 0.90973532,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.931674,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.659853219985962
    },
    {
      "auxiliary_loss_clip": 0.01160304,
      "auxiliary_loss_mlp": 0.01021411,
      "balance_loss_clip": 1.00761962,
      "balance_loss_mlp": 1.01431489,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.925556423665575,
      "language_loss": 0.74856079,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77037793,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.6117734909057617
    },
    {
      "auxiliary_loss_clip": 0.01155226,
      "auxiliary_loss_mlp": 0.01024127,
      "balance_loss_clip": 0.93082631,
      "balance_loss_mlp": 1.01691508,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 1.7297168489196075,
      "language_loss": 0.83079851,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.85259205,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.671342611312866
    },
    {
      "auxiliary_loss_clip": 0.01167387,
      "auxiliary_loss_mlp": 0.01030612,
      "balance_loss_clip": 1.00984967,
      "balance_loss_mlp": 1.02353084,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 1.8939569863692451,
      "language_loss": 0.76079309,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78277302,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 2.6226933002471924
    },
    {
      "auxiliary_loss_clip": 0.01158456,
      "auxiliary_loss_mlp": 0.01027029,
      "balance_loss_clip": 0.93212783,
      "balance_loss_mlp": 1.02041852,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 2.3668705876670204,
      "language_loss": 0.74647349,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.76832831,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.6877613067626953
    },
    {
      "auxiliary_loss_clip": 0.01175078,
      "auxiliary_loss_mlp": 0.01024515,
      "balance_loss_clip": 0.85785848,
      "balance_loss_mlp": 1.01680839,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.314177964251473,
      "language_loss": 0.84193611,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.86393201,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 2.9695184230804443
    },
    {
      "auxiliary_loss_clip": 0.01158351,
      "auxiliary_loss_mlp": 0.01025381,
      "balance_loss_clip": 0.9708581,
      "balance_loss_mlp": 1.01869607,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 1.9329839060839356,
      "language_loss": 0.8044287,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.82626605,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.8660106658935547
    },
    {
      "auxiliary_loss_clip": 0.01155802,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 0.96649945,
      "balance_loss_mlp": 1.01986539,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 1.687585492393405,
      "language_loss": 0.85788035,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.87970567,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 3.536728858947754
    },
    {
      "auxiliary_loss_clip": 0.01168261,
      "auxiliary_loss_mlp": 0.01022276,
      "balance_loss_clip": 1.04773927,
      "balance_loss_mlp": 1.01505446,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 1.6867661359590258,
      "language_loss": 0.63062489,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65253025,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.6530635356903076
    },
    {
      "auxiliary_loss_clip": 0.01156726,
      "auxiliary_loss_mlp": 0.01025079,
      "balance_loss_clip": 0.93079537,
      "balance_loss_mlp": 1.01820087,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 1.6437865490103782,
      "language_loss": 0.69160771,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71342576,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.6938042640686035
    },
    {
      "auxiliary_loss_clip": 0.01162153,
      "auxiliary_loss_mlp": 0.01023346,
      "balance_loss_clip": 0.96920693,
      "balance_loss_mlp": 1.01558506,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 1.9718654668842777,
      "language_loss": 0.74331677,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.76517177,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.7236268520355225
    },
    {
      "auxiliary_loss_clip": 0.01167942,
      "auxiliary_loss_mlp": 0.01027727,
      "balance_loss_clip": 1.00883031,
      "balance_loss_mlp": 1.02165353,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.5485063245261432,
      "language_loss": 0.82159245,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84354919,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.7069740295410156
    },
    {
      "auxiliary_loss_clip": 0.01165818,
      "auxiliary_loss_mlp": 0.01025118,
      "balance_loss_clip": 1.01061416,
      "balance_loss_mlp": 1.01791191,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 1.5393190618164507,
      "language_loss": 0.76182806,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78373742,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 3.6962008476257324
    },
    {
      "auxiliary_loss_clip": 0.01166953,
      "auxiliary_loss_mlp": 0.01025805,
      "balance_loss_clip": 1.04930508,
      "balance_loss_mlp": 1.01887262,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 1.859724112403196,
      "language_loss": 0.86050856,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.8824361,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 3.4419164657592773
    },
    {
      "auxiliary_loss_clip": 0.01166935,
      "auxiliary_loss_mlp": 0.01031885,
      "balance_loss_clip": 1.01024687,
      "balance_loss_mlp": 1.02515531,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.5470347656865153,
      "language_loss": 0.79118133,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81316948,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 3.71828556060791
    },
    {
      "auxiliary_loss_clip": 0.01169158,
      "auxiliary_loss_mlp": 0.01022948,
      "balance_loss_clip": 1.04927373,
      "balance_loss_mlp": 1.01584888,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 1.848838763052677,
      "language_loss": 0.85165846,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.8735795,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 2.5436699390411377
    },
    {
      "auxiliary_loss_clip": 0.01162927,
      "auxiliary_loss_mlp": 0.01022788,
      "balance_loss_clip": 0.93212974,
      "balance_loss_mlp": 1.01586461,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 2.157980610212242,
      "language_loss": 0.68909556,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.71095276,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 2.7506792545318604
    },
    {
      "auxiliary_loss_clip": 0.01156308,
      "auxiliary_loss_mlp": 0.01024764,
      "balance_loss_clip": 0.96591276,
      "balance_loss_mlp": 1.01773643,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 1.893466044278317,
      "language_loss": 0.79357028,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81538099,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.58796763420105
    },
    {
      "auxiliary_loss_clip": 0.0116293,
      "auxiliary_loss_mlp": 0.01022193,
      "balance_loss_clip": 0.96758687,
      "balance_loss_mlp": 1.01551437,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.9792935945376913,
      "language_loss": 0.84761083,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.86946207,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 2.7251839637756348
    },
    {
      "auxiliary_loss_clip": 0.01165529,
      "auxiliary_loss_mlp": 0.01032117,
      "balance_loss_clip": 0.97006404,
      "balance_loss_mlp": 1.02431464,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 1.775199905180169,
      "language_loss": 0.68975258,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.71172899,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 2.6325395107269287
    },
    {
      "auxiliary_loss_clip": 0.01160031,
      "auxiliary_loss_mlp": 0.01019868,
      "balance_loss_clip": 1.00805962,
      "balance_loss_mlp": 1.01312971,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 2.310542087278874,
      "language_loss": 0.74878132,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.77058041,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.658562421798706
    },
    {
      "auxiliary_loss_clip": 0.01168612,
      "auxiliary_loss_mlp": 0.01031421,
      "balance_loss_clip": 1.04888141,
      "balance_loss_mlp": 1.02386594,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 2.312291611508791,
      "language_loss": 0.6852448,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.70724511,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 2.648514747619629
    },
    {
      "auxiliary_loss_clip": 0.0116112,
      "auxiliary_loss_mlp": 0.01021251,
      "balance_loss_clip": 0.97087055,
      "balance_loss_mlp": 1.01489449,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 1.6532937724078773,
      "language_loss": 0.74930906,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.77113283,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 2.6410388946533203
    },
    {
      "auxiliary_loss_clip": 0.01162805,
      "auxiliary_loss_mlp": 0.01122093,
      "balance_loss_clip": 1.01059222,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 1.6027721060267353,
      "language_loss": 0.79424727,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.81709617,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.591984510421753
    },
    {
      "auxiliary_loss_clip": 0.01164569,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.00989354,
      "balance_loss_mlp": 1.02039468,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 1.676897680770456,
      "language_loss": 0.66121876,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.68314296,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 2.6140687465667725
    },
    {
      "auxiliary_loss_clip": 0.01156921,
      "auxiliary_loss_mlp": 0.01026349,
      "balance_loss_clip": 0.93285424,
      "balance_loss_mlp": 1.01952136,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.9031488776609622,
      "language_loss": 0.79394907,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81578177,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.660292148590088
    },
    {
      "auxiliary_loss_clip": 0.01154878,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 0.96672082,
      "balance_loss_mlp": 1.01978362,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 1.8967227964363877,
      "language_loss": 0.67567748,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69748771,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.6271867752075195
    },
    {
      "auxiliary_loss_clip": 0.01064123,
      "auxiliary_loss_mlp": 0.01001944,
      "balance_loss_clip": 0.93489718,
      "balance_loss_mlp": 1.00026274,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7913127780848326,
      "language_loss": 0.58698756,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.60764819,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.2305097579956055
    },
    {
      "auxiliary_loss_clip": 0.0115556,
      "auxiliary_loss_mlp": 0.01034594,
      "balance_loss_clip": 0.97089064,
      "balance_loss_mlp": 1.02717316,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 2.4827274154529815,
      "language_loss": 0.71994978,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.74185133,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.6649961471557617
    },
    {
      "auxiliary_loss_clip": 0.01168199,
      "auxiliary_loss_mlp": 0.01030452,
      "balance_loss_clip": 1.00884235,
      "balance_loss_mlp": 1.02372861,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 1.8419000149762543,
      "language_loss": 0.75503027,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.77701688,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.5916740894317627
    },
    {
      "auxiliary_loss_clip": 0.01164507,
      "auxiliary_loss_mlp": 0.01025911,
      "balance_loss_clip": 0.97066694,
      "balance_loss_mlp": 1.01900315,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 2.248879808332816,
      "language_loss": 0.8058995,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82780373,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.6717753410339355
    },
    {
      "auxiliary_loss_clip": 0.01166701,
      "auxiliary_loss_mlp": 0.01029009,
      "balance_loss_clip": 1.04653788,
      "balance_loss_mlp": 1.02145672,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 2.5393897764823583,
      "language_loss": 0.68781656,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.70977366,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.535511016845703
    },
    {
      "auxiliary_loss_clip": 0.01165529,
      "auxiliary_loss_mlp": 0.01022815,
      "balance_loss_clip": 0.93292361,
      "balance_loss_mlp": 1.01616323,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 1.6578320278223078,
      "language_loss": 0.78684658,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80873001,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 3.617368221282959
    },
    {
      "auxiliary_loss_clip": 0.01156622,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 0.93144029,
      "balance_loss_mlp": 1.02277136,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.3960087185850083,
      "language_loss": 0.73486209,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.7567265,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.7134695053100586
    },
    {
      "auxiliary_loss_clip": 0.01163922,
      "auxiliary_loss_mlp": 0.01024581,
      "balance_loss_clip": 0.97192973,
      "balance_loss_mlp": 1.01780701,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.6421447687808701,
      "language_loss": 0.78503817,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80692315,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.6684987545013428
    },
    {
      "auxiliary_loss_clip": 0.01162561,
      "auxiliary_loss_mlp": 0.01022831,
      "balance_loss_clip": 0.97298902,
      "balance_loss_mlp": 1.01573241,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.2482173754882293,
      "language_loss": 0.78655416,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80840808,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.722346544265747
    },
    {
      "auxiliary_loss_clip": 0.01164323,
      "auxiliary_loss_mlp": 0.01019525,
      "balance_loss_clip": 0.93328369,
      "balance_loss_mlp": 1.01297116,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 2.657234319924327,
      "language_loss": 0.83145273,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.85329115,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.631204843521118
    },
    {
      "auxiliary_loss_clip": 0.0116908,
      "auxiliary_loss_mlp": 0.0102352,
      "balance_loss_clip": 0.97312099,
      "balance_loss_mlp": 1.01605701,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 2.1901745007975277,
      "language_loss": 0.75755161,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.7794776,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.714148998260498
    },
    {
      "auxiliary_loss_clip": 0.01155666,
      "auxiliary_loss_mlp": 0.01026831,
      "balance_loss_clip": 0.9657169,
      "balance_loss_mlp": 1.01925778,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 2.893274401770568,
      "language_loss": 0.56776273,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.58958769,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 3.619877576828003
    },
    {
      "auxiliary_loss_clip": 0.01172188,
      "auxiliary_loss_mlp": 0.01021431,
      "balance_loss_clip": 0.97166985,
      "balance_loss_mlp": 1.01402235,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 1.7256332685721625,
      "language_loss": 0.76841748,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.79035366,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 4.490628719329834
    },
    {
      "auxiliary_loss_clip": 0.01163225,
      "auxiliary_loss_mlp": 0.01025811,
      "balance_loss_clip": 0.8180607,
      "balance_loss_mlp": 1.01807737,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.53499978054005,
      "language_loss": 0.82093811,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.84282845,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.8585331439971924
    },
    {
      "auxiliary_loss_clip": 0.01157942,
      "auxiliary_loss_mlp": 0.01022122,
      "balance_loss_clip": 0.9664917,
      "balance_loss_mlp": 1.01510906,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 2.032001297343819,
      "language_loss": 0.71042955,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.73223019,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 2.805222511291504
    },
    {
      "auxiliary_loss_clip": 0.01158718,
      "auxiliary_loss_mlp": 0.01027686,
      "balance_loss_clip": 0.85196114,
      "balance_loss_mlp": 1.02051544,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.1997457793782482,
      "language_loss": 0.74632466,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.76818871,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 2.697401285171509
    },
    {
      "auxiliary_loss_clip": 0.01169884,
      "auxiliary_loss_mlp": 0.01028609,
      "balance_loss_clip": 1.00899088,
      "balance_loss_mlp": 1.02087784,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 2.5243476018363684,
      "language_loss": 0.76928449,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.79126942,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.6767995357513428
    },
    {
      "auxiliary_loss_clip": 0.01169859,
      "auxiliary_loss_mlp": 0.01026195,
      "balance_loss_clip": 1.00765765,
      "balance_loss_mlp": 1.01875019,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.151774015542573,
      "language_loss": 0.69329512,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71525562,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 2.589154005050659
    },
    {
      "auxiliary_loss_clip": 0.01169013,
      "auxiliary_loss_mlp": 0.01026562,
      "balance_loss_clip": 1.00876856,
      "balance_loss_mlp": 1.01967776,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 1.802032363465531,
      "language_loss": 0.76165199,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.78360772,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 2.6839101314544678
    },
    {
      "auxiliary_loss_clip": 0.01166727,
      "auxiliary_loss_mlp": 0.01029962,
      "balance_loss_clip": 0.93590391,
      "balance_loss_mlp": 1.02359939,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 1.841980431488164,
      "language_loss": 0.68133789,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70330477,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 2.6395366191864014
    },
    {
      "auxiliary_loss_clip": 0.01073147,
      "auxiliary_loss_mlp": 0.01002248,
      "balance_loss_clip": 0.86084366,
      "balance_loss_mlp": 1.00045967,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.771108848488092,
      "language_loss": 0.54515898,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56591296,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.346752166748047
    },
    {
      "auxiliary_loss_clip": 0.01165354,
      "auxiliary_loss_mlp": 0.01025806,
      "balance_loss_clip": 1.00961566,
      "balance_loss_mlp": 1.01899362,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.40397351362345,
      "language_loss": 0.73776251,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.75967413,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 2.571418046951294
    },
    {
      "auxiliary_loss_clip": 0.01160508,
      "auxiliary_loss_mlp": 0.01025071,
      "balance_loss_clip": 1.00893092,
      "balance_loss_mlp": 1.01827633,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.623367604195172,
      "language_loss": 0.79253101,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81438684,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.664802074432373
    },
    {
      "auxiliary_loss_clip": 0.01066107,
      "auxiliary_loss_mlp": 0.01001214,
      "balance_loss_clip": 0.93371534,
      "balance_loss_mlp": 0.9996047,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7185076120060846,
      "language_loss": 0.56810313,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58877629,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 3.0635218620300293
    },
    {
      "auxiliary_loss_clip": 0.01162595,
      "auxiliary_loss_mlp": 0.01026173,
      "balance_loss_clip": 1.01019645,
      "balance_loss_mlp": 1.01946473,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 1.6274213223516025,
      "language_loss": 0.76487637,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78676403,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.6470046043395996
    },
    {
      "auxiliary_loss_clip": 0.01156534,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 0.96945906,
      "balance_loss_mlp": 1.02056384,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 1.4709034783754549,
      "language_loss": 0.70769745,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.72953957,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.718832492828369
    },
    {
      "auxiliary_loss_clip": 0.0115561,
      "auxiliary_loss_mlp": 0.01027362,
      "balance_loss_clip": 0.96594739,
      "balance_loss_mlp": 1.02028704,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 1.5972398173785265,
      "language_loss": 0.81317729,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83500695,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.5853428840637207
    },
    {
      "auxiliary_loss_clip": 0.01167561,
      "auxiliary_loss_mlp": 0.01021039,
      "balance_loss_clip": 1.04567575,
      "balance_loss_mlp": 1.01417303,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 1.8155191927928451,
      "language_loss": 0.78677654,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.80866253,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.56266188621521
    },
    {
      "auxiliary_loss_clip": 0.01162958,
      "auxiliary_loss_mlp": 0.01023805,
      "balance_loss_clip": 0.96937364,
      "balance_loss_mlp": 1.01723361,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 1.9067061551064632,
      "language_loss": 0.75322509,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77509272,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.5842843055725098
    },
    {
      "auxiliary_loss_clip": 0.01163258,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 0.93444026,
      "balance_loss_mlp": 1.02121758,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 1.9479395993084998,
      "language_loss": 0.76866841,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.79058063,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.7916672229766846
    },
    {
      "auxiliary_loss_clip": 0.01159014,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 0.93050909,
      "balance_loss_mlp": 1.01767969,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 2.0576683165083787,
      "language_loss": 0.68360639,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70544213,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.7148497104644775
    },
    {
      "auxiliary_loss_clip": 0.01161523,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 0.96895105,
      "balance_loss_mlp": 1.01712966,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 1.6128489399543309,
      "language_loss": 0.75298244,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77484775,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 3.775993824005127
    },
    {
      "auxiliary_loss_clip": 0.01162155,
      "auxiliary_loss_mlp": 0.01019422,
      "balance_loss_clip": 0.9725194,
      "balance_loss_mlp": 1.01294255,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.5972685293037732,
      "language_loss": 0.79779273,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.81960851,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.720743417739868
    },
    {
      "auxiliary_loss_clip": 0.01166983,
      "auxiliary_loss_mlp": 0.01022561,
      "balance_loss_clip": 1.04980636,
      "balance_loss_mlp": 1.01597726,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 1.9695610909169163,
      "language_loss": 0.7965377,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.81843311,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.55644154548645
    },
    {
      "auxiliary_loss_clip": 0.0114958,
      "auxiliary_loss_mlp": 0.01022045,
      "balance_loss_clip": 0.85187316,
      "balance_loss_mlp": 1.01531267,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.5615553140820069,
      "language_loss": 0.61515385,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.63687009,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.8373003005981445
    },
    {
      "auxiliary_loss_clip": 0.01155125,
      "auxiliary_loss_mlp": 0.0102573,
      "balance_loss_clip": 0.81746602,
      "balance_loss_mlp": 1.01890576,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.4969747852978657,
      "language_loss": 0.69055212,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.71236062,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.8915839195251465
    },
    {
      "auxiliary_loss_clip": 0.01163071,
      "auxiliary_loss_mlp": 0.01023695,
      "balance_loss_clip": 1.0095489,
      "balance_loss_mlp": 1.01670051,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 1.9404717097940904,
      "language_loss": 0.72071368,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.74258137,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 3.0550284385681152
    },
    {
      "auxiliary_loss_clip": 0.01070873,
      "auxiliary_loss_mlp": 0.01005477,
      "balance_loss_clip": 0.86027658,
      "balance_loss_mlp": 1.00384426,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.8158681166581297,
      "language_loss": 0.58122361,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60198712,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 4.329629421234131
    },
    {
      "auxiliary_loss_clip": 0.01159949,
      "auxiliary_loss_mlp": 0.01122706,
      "balance_loss_clip": 0.89524472,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 1.8832517808951117,
      "language_loss": 0.60199833,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.62482488,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 3.9973416328430176
    },
    {
      "auxiliary_loss_clip": 0.01163131,
      "auxiliary_loss_mlp": 0.01030847,
      "balance_loss_clip": 0.93197286,
      "balance_loss_mlp": 1.02407312,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.1132790531802796,
      "language_loss": 0.66850978,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.6904496,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.6747422218322754
    },
    {
      "auxiliary_loss_clip": 0.01164831,
      "auxiliary_loss_mlp": 0.01033856,
      "balance_loss_clip": 0.9715721,
      "balance_loss_mlp": 1.02648902,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 2.5397332308845715,
      "language_loss": 0.68308443,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.70507133,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.6651177406311035
    },
    {
      "auxiliary_loss_clip": 0.01164961,
      "auxiliary_loss_mlp": 0.0102308,
      "balance_loss_clip": 0.93464327,
      "balance_loss_mlp": 1.01624382,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 1.6484309581346024,
      "language_loss": 0.82706821,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.8489486,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 2.706915855407715
    },
    {
      "auxiliary_loss_clip": 0.01165715,
      "auxiliary_loss_mlp": 0.01122106,
      "balance_loss_clip": 1.01173735,
      "balance_loss_mlp": 0.0,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 3.507756129179172,
      "language_loss": 0.75816625,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.78104448,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.6351802349090576
    },
    {
      "auxiliary_loss_clip": 0.0115486,
      "auxiliary_loss_mlp": 0.01022103,
      "balance_loss_clip": 0.89036405,
      "balance_loss_mlp": 1.01550794,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 1.9125936580269745,
      "language_loss": 0.77021819,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79198784,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 2.6703505516052246
    },
    {
      "auxiliary_loss_clip": 0.01169135,
      "auxiliary_loss_mlp": 0.01028189,
      "balance_loss_clip": 1.01080596,
      "balance_loss_mlp": 1.02105165,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 1.943936010331348,
      "language_loss": 0.63987005,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.6618433,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 2.6124236583709717
    },
    {
      "auxiliary_loss_clip": 0.01159889,
      "auxiliary_loss_mlp": 0.01025377,
      "balance_loss_clip": 0.96910655,
      "balance_loss_mlp": 1.01845098,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 2.632415004186641,
      "language_loss": 0.6182158,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.64006841,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 2.723388195037842
    },
    {
      "auxiliary_loss_clip": 0.01160576,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 0.96849924,
      "balance_loss_mlp": 1.01875782,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 1.8930002267901536,
      "language_loss": 0.71843374,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74030018,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 2.6724791526794434
    },
    {
      "auxiliary_loss_clip": 0.01167468,
      "auxiliary_loss_mlp": 0.01022842,
      "balance_loss_clip": 1.00901818,
      "balance_loss_mlp": 1.01600838,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 1.774614622687301,
      "language_loss": 0.68903184,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.710935,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.6295223236083984
    },
    {
      "auxiliary_loss_clip": 0.01159866,
      "auxiliary_loss_mlp": 0.01030571,
      "balance_loss_clip": 1.00767636,
      "balance_loss_mlp": 1.02324009,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 1.6178470571034165,
      "language_loss": 0.76499957,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.78690392,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 2.5897202491760254
    },
    {
      "auxiliary_loss_clip": 0.01166906,
      "auxiliary_loss_mlp": 0.01022438,
      "balance_loss_clip": 0.97221863,
      "balance_loss_mlp": 1.01556277,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.7138797935211256,
      "language_loss": 0.86485535,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88674879,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.6618404388427734
    },
    {
      "auxiliary_loss_clip": 0.01159672,
      "auxiliary_loss_mlp": 0.01028855,
      "balance_loss_clip": 0.85648882,
      "balance_loss_mlp": 1.02211642,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 7.808679699993672,
      "language_loss": 0.67840552,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.7002908,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.7636830806732178
    },
    {
      "auxiliary_loss_clip": 0.01163882,
      "auxiliary_loss_mlp": 0.0102121,
      "balance_loss_clip": 1.00874567,
      "balance_loss_mlp": 1.01517463,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 2.011383216703443,
      "language_loss": 0.76066536,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78251624,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 2.6139039993286133
    },
    {
      "auxiliary_loss_clip": 0.0116122,
      "auxiliary_loss_mlp": 0.01030093,
      "balance_loss_clip": 0.93406284,
      "balance_loss_mlp": 1.02329254,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 1.588119327272424,
      "language_loss": 0.79280478,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81471789,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.6776206493377686
    },
    {
      "auxiliary_loss_clip": 0.0116359,
      "auxiliary_loss_mlp": 0.01022709,
      "balance_loss_clip": 1.00599504,
      "balance_loss_mlp": 1.01577687,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 1.808923350257961,
      "language_loss": 0.75194132,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77380425,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.6313328742980957
    },
    {
      "auxiliary_loss_clip": 0.01168462,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 0.97438967,
      "balance_loss_mlp": 1.02263975,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.710918083294948,
      "language_loss": 0.74521774,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.767205,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.6454882621765137
    },
    {
      "auxiliary_loss_clip": 0.01161577,
      "auxiliary_loss_mlp": 0.0102274,
      "balance_loss_clip": 0.89374888,
      "balance_loss_mlp": 1.01609969,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 2.0291095461692765,
      "language_loss": 0.7293241,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.75116724,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 2.690842628479004
    },
    {
      "auxiliary_loss_clip": 0.01169919,
      "auxiliary_loss_mlp": 0.01122111,
      "balance_loss_clip": 0.97378117,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 2.1083745803517058,
      "language_loss": 0.77644312,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.79936343,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.6103758811950684
    },
    {
      "auxiliary_loss_clip": 0.01167509,
      "auxiliary_loss_mlp": 0.01025603,
      "balance_loss_clip": 0.93192911,
      "balance_loss_mlp": 1.01850426,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.474447519483,
      "language_loss": 0.7514137,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.77334487,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.699050188064575
    },
    {
      "auxiliary_loss_clip": 0.01158017,
      "auxiliary_loss_mlp": 0.01022471,
      "balance_loss_clip": 0.96691078,
      "balance_loss_mlp": 1.0149281,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 1.866671214380235,
      "language_loss": 0.57171643,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59352136,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 3.530846357345581
    },
    {
      "auxiliary_loss_clip": 0.01066937,
      "auxiliary_loss_mlp": 0.01001596,
      "balance_loss_clip": 0.93404943,
      "balance_loss_mlp": 1.00002217,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6382333058459316,
      "language_loss": 0.51676798,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53745329,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.318838357925415
    },
    {
      "auxiliary_loss_clip": 0.01163958,
      "auxiliary_loss_mlp": 0.01022587,
      "balance_loss_clip": 1.01024723,
      "balance_loss_mlp": 1.01525593,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 1.7705636075481759,
      "language_loss": 0.88614154,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90800691,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.6570215225219727
    },
    {
      "auxiliary_loss_clip": 0.01168088,
      "auxiliary_loss_mlp": 0.01122405,
      "balance_loss_clip": 0.9315058,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.080336021812658,
      "language_loss": 0.78294653,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.80585146,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.748180866241455
    },
    {
      "auxiliary_loss_clip": 0.01163244,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 0.93370575,
      "balance_loss_mlp": 1.01874018,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.1293473730382315,
      "language_loss": 0.66608065,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.68796605,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.713223457336426
    },
    {
      "auxiliary_loss_clip": 0.01159536,
      "auxiliary_loss_mlp": 0.01025027,
      "balance_loss_clip": 0.93321609,
      "balance_loss_mlp": 1.01730251,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 1.6691423918796453,
      "language_loss": 0.68318951,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70503521,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 3.5192737579345703
    },
    {
      "auxiliary_loss_clip": 0.01155979,
      "auxiliary_loss_mlp": 0.01023695,
      "balance_loss_clip": 0.97067803,
      "balance_loss_mlp": 1.01738274,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 1.8070561625982906,
      "language_loss": 0.80167657,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.82347327,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 3.61673903465271
    },
    {
      "auxiliary_loss_clip": 0.01155767,
      "auxiliary_loss_mlp": 0.01028869,
      "balance_loss_clip": 0.9363392,
      "balance_loss_mlp": 1.02223539,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.4331885808209637,
      "language_loss": 0.68414944,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.7059958,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 3.796963930130005
    },
    {
      "auxiliary_loss_clip": 0.01160637,
      "auxiliary_loss_mlp": 0.01028921,
      "balance_loss_clip": 0.8555603,
      "balance_loss_mlp": 1.02196229,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 1.7371044603878278,
      "language_loss": 0.77721763,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.79911321,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.747098207473755
    },
    {
      "auxiliary_loss_clip": 0.01162432,
      "auxiliary_loss_mlp": 0.01024019,
      "balance_loss_clip": 0.96967018,
      "balance_loss_mlp": 1.0176084,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.9125694018209434,
      "language_loss": 0.75774151,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.77960604,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.6672568321228027
    },
    {
      "auxiliary_loss_clip": 0.01172127,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.05149174,
      "balance_loss_mlp": 1.02051401,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 4.324125356409329,
      "language_loss": 0.64096189,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.66296238,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 2.5501744747161865
    },
    {
      "auxiliary_loss_clip": 0.01164339,
      "auxiliary_loss_mlp": 0.01122088,
      "balance_loss_clip": 1.00713944,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 2.0699707220207784,
      "language_loss": 0.55520713,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.57807142,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 2.5819737911224365
    },
    {
      "auxiliary_loss_clip": 0.0116698,
      "auxiliary_loss_mlp": 0.0102654,
      "balance_loss_clip": 1.01016068,
      "balance_loss_mlp": 1.01933992,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 1.9208047836821491,
      "language_loss": 0.74053359,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76246881,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 2.626762866973877
    },
    {
      "auxiliary_loss_clip": 0.01153589,
      "auxiliary_loss_mlp": 0.01028636,
      "balance_loss_clip": 1.00614142,
      "balance_loss_mlp": 1.02156067,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 1.8729994940171835,
      "language_loss": 0.76131725,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.78313947,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 2.6202352046966553
    },
    {
      "auxiliary_loss_clip": 0.01157148,
      "auxiliary_loss_mlp": 0.01020983,
      "balance_loss_clip": 0.96889257,
      "balance_loss_mlp": 1.01398516,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.6801051523527664,
      "language_loss": 0.65328991,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.67507124,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 2.6351828575134277
    },
    {
      "auxiliary_loss_clip": 0.01157906,
      "auxiliary_loss_mlp": 0.01122038,
      "balance_loss_clip": 0.89276791,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.113964980111409,
      "language_loss": 0.77539283,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.79819232,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.696394205093384
    },
    {
      "auxiliary_loss_clip": 0.01169721,
      "auxiliary_loss_mlp": 0.01026888,
      "balance_loss_clip": 0.93222594,
      "balance_loss_mlp": 1.0190444,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 1.9241140277398836,
      "language_loss": 0.7578811,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.7798472,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 2.633875846862793
    },
    {
      "auxiliary_loss_clip": 0.01163305,
      "auxiliary_loss_mlp": 0.010249,
      "balance_loss_clip": 0.93253881,
      "balance_loss_mlp": 1.01752126,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 1.822902734949673,
      "language_loss": 0.68175679,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70363879,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.7068533897399902
    },
    {
      "auxiliary_loss_clip": 0.01167682,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.04767454,
      "balance_loss_mlp": 1.02192187,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 1.9936136032342193,
      "language_loss": 0.68645227,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70842081,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.9060235023498535
    },
    {
      "auxiliary_loss_clip": 0.01165345,
      "auxiliary_loss_mlp": 0.01022597,
      "balance_loss_clip": 1.01150537,
      "balance_loss_mlp": 1.01518202,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 2.475599096586741,
      "language_loss": 0.83714229,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85902172,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 2.5657248497009277
    },
    {
      "auxiliary_loss_clip": 0.01166113,
      "auxiliary_loss_mlp": 0.01026304,
      "balance_loss_clip": 1.00750256,
      "balance_loss_mlp": 1.01921988,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 1.7939546841571445,
      "language_loss": 0.81933236,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.8412565,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.6514134407043457
    },
    {
      "auxiliary_loss_clip": 0.01164839,
      "auxiliary_loss_mlp": 0.01122238,
      "balance_loss_clip": 1.00864911,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 1.8159979654339822,
      "language_loss": 0.73718518,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.76005602,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.6358754634857178
    },
    {
      "auxiliary_loss_clip": 0.01156162,
      "auxiliary_loss_mlp": 0.01028279,
      "balance_loss_clip": 0.96993351,
      "balance_loss_mlp": 1.02126408,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 1.7700208416450287,
      "language_loss": 0.77587068,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79771513,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.6368179321289062
    },
    {
      "auxiliary_loss_clip": 0.01168646,
      "auxiliary_loss_mlp": 0.01027347,
      "balance_loss_clip": 1.01429033,
      "balance_loss_mlp": 1.0202179,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 1.6047924813074832,
      "language_loss": 0.75799525,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.77995515,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.5879054069519043
    },
    {
      "auxiliary_loss_clip": 0.01164061,
      "auxiliary_loss_mlp": 0.01032351,
      "balance_loss_clip": 1.00946939,
      "balance_loss_mlp": 1.02501988,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 2.25285781033494,
      "language_loss": 0.71514195,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.73710608,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 2.604403495788574
    },
    {
      "auxiliary_loss_clip": 0.01173009,
      "auxiliary_loss_mlp": 0.01027072,
      "balance_loss_clip": 0.97288465,
      "balance_loss_mlp": 1.02014613,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 2.1939854752324255,
      "language_loss": 0.66548753,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.68748832,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 3.5555689334869385
    },
    {
      "auxiliary_loss_clip": 0.01168581,
      "auxiliary_loss_mlp": 0.01029818,
      "balance_loss_clip": 0.97128296,
      "balance_loss_mlp": 1.02311504,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 3.708931348022523,
      "language_loss": 0.79975879,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.82174277,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.656870126724243
    },
    {
      "auxiliary_loss_clip": 0.01058789,
      "auxiliary_loss_mlp": 0.01001363,
      "balance_loss_clip": 1.00795996,
      "balance_loss_mlp": 0.99977702,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7833842277656571,
      "language_loss": 0.60105145,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62165296,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.242858409881592
    },
    {
      "auxiliary_loss_clip": 0.01164719,
      "auxiliary_loss_mlp": 0.01026566,
      "balance_loss_clip": 1.01025414,
      "balance_loss_mlp": 1.01988769,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 2.0715164212268333,
      "language_loss": 0.73430872,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.75622153,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.545348644256592
    },
    {
      "auxiliary_loss_clip": 0.01164494,
      "auxiliary_loss_mlp": 0.01026234,
      "balance_loss_clip": 0.97047132,
      "balance_loss_mlp": 1.01905215,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 2.1417865666632006,
      "language_loss": 0.79972547,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.8216328,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.7256603240966797
    },
    {
      "auxiliary_loss_clip": 0.01164282,
      "auxiliary_loss_mlp": 0.01023577,
      "balance_loss_clip": 1.04821515,
      "balance_loss_mlp": 1.01725304,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.383711951565977,
      "language_loss": 0.73253727,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75441587,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.522993803024292
    },
    {
      "auxiliary_loss_clip": 0.01170138,
      "auxiliary_loss_mlp": 0.01031661,
      "balance_loss_clip": 1.01185715,
      "balance_loss_mlp": 1.02436268,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 1.9811642941040155,
      "language_loss": 0.79792261,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81994069,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 3.6599960327148438
    },
    {
      "auxiliary_loss_clip": 0.01165747,
      "auxiliary_loss_mlp": 0.0103076,
      "balance_loss_clip": 0.89554286,
      "balance_loss_mlp": 1.0231427,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 2.564954420843524,
      "language_loss": 0.66225898,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68422413,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.7261741161346436
    },
    {
      "auxiliary_loss_clip": 0.01163139,
      "auxiliary_loss_mlp": 0.01022775,
      "balance_loss_clip": 1.0047555,
      "balance_loss_mlp": 1.01580477,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 1.79275889038004,
      "language_loss": 0.70182729,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72368646,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 3.5480384826660156
    },
    {
      "auxiliary_loss_clip": 0.01167678,
      "auxiliary_loss_mlp": 0.01031066,
      "balance_loss_clip": 1.01251888,
      "balance_loss_mlp": 1.02427745,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 1.6987812989376085,
      "language_loss": 0.80766892,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.82965636,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 3.5026586055755615
    },
    {
      "auxiliary_loss_clip": 0.01155467,
      "auxiliary_loss_mlp": 0.01024088,
      "balance_loss_clip": 0.97001255,
      "balance_loss_mlp": 1.01707518,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 2.76315090909936,
      "language_loss": 0.72439456,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.74619007,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.6600470542907715
    },
    {
      "auxiliary_loss_clip": 0.01168325,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.04706383,
      "balance_loss_mlp": 1.02126122,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 2.1528585984511386,
      "language_loss": 0.70211697,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72408611,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 2.6324424743652344
    },
    {
      "auxiliary_loss_clip": 0.01164823,
      "auxiliary_loss_mlp": 0.0102769,
      "balance_loss_clip": 0.93625891,
      "balance_loss_mlp": 1.02026367,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.5684865346826076,
      "language_loss": 0.73424131,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75616646,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.731877088546753
    },
    {
      "auxiliary_loss_clip": 0.0115886,
      "auxiliary_loss_mlp": 0.01025438,
      "balance_loss_clip": 0.89648032,
      "balance_loss_mlp": 1.01806819,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.0730691025197934,
      "language_loss": 0.77848846,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.80033141,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 2.67396879196167
    },
    {
      "auxiliary_loss_clip": 0.01161956,
      "auxiliary_loss_mlp": 0.01022874,
      "balance_loss_clip": 1.00743639,
      "balance_loss_mlp": 1.01643419,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.48898281926179,
      "language_loss": 0.7153275,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73717582,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 2.686976432800293
    },
    {
      "auxiliary_loss_clip": 0.01160455,
      "auxiliary_loss_mlp": 0.01022511,
      "balance_loss_clip": 0.97067201,
      "balance_loss_mlp": 1.01571608,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 2.2298018923821417,
      "language_loss": 0.67590714,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69773686,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.695977210998535
    },
    {
      "auxiliary_loss_clip": 0.01156458,
      "auxiliary_loss_mlp": 0.01026401,
      "balance_loss_clip": 0.89380455,
      "balance_loss_mlp": 1.01890516,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 2.001764279672521,
      "language_loss": 0.71683496,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.73866355,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 2.7783243656158447
    },
    {
      "auxiliary_loss_clip": 0.01163221,
      "auxiliary_loss_mlp": 0.01023765,
      "balance_loss_clip": 0.97132343,
      "balance_loss_mlp": 1.01726246,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 2.447758844672781,
      "language_loss": 0.68122137,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70309126,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 2.6308422088623047
    },
    {
      "auxiliary_loss_clip": 0.01162058,
      "auxiliary_loss_mlp": 0.01026477,
      "balance_loss_clip": 0.96984899,
      "balance_loss_mlp": 1.01894605,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 2.298237974173595,
      "language_loss": 0.70159686,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72348225,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 2.663376808166504
    },
    {
      "auxiliary_loss_clip": 0.01166006,
      "auxiliary_loss_mlp": 0.01020284,
      "balance_loss_clip": 0.93785954,
      "balance_loss_mlp": 1.01393652,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 1.7068739769347843,
      "language_loss": 0.87312734,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89499027,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.7077057361602783
    },
    {
      "auxiliary_loss_clip": 0.01160621,
      "auxiliary_loss_mlp": 0.01025559,
      "balance_loss_clip": 1.00457811,
      "balance_loss_mlp": 1.01918983,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 1.8259899611520571,
      "language_loss": 0.76820052,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79006231,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.669081687927246
    },
    {
      "auxiliary_loss_clip": 0.01165882,
      "auxiliary_loss_mlp": 0.01026698,
      "balance_loss_clip": 1.04838562,
      "balance_loss_mlp": 1.01952195,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 1.7352058202011775,
      "language_loss": 0.76726437,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.78919017,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 2.7262308597564697
    },
    {
      "auxiliary_loss_clip": 0.01155932,
      "auxiliary_loss_mlp": 0.01026133,
      "balance_loss_clip": 1.00705612,
      "balance_loss_mlp": 1.01903701,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.0795507389191306,
      "language_loss": 0.74016583,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76198649,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.756380319595337
    },
    {
      "auxiliary_loss_clip": 0.01156688,
      "auxiliary_loss_mlp": 0.01026872,
      "balance_loss_clip": 1.00585437,
      "balance_loss_mlp": 1.02035713,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 1.889471976185711,
      "language_loss": 0.71539783,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.7372334,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.6505868434906006
    },
    {
      "auxiliary_loss_clip": 0.01166436,
      "auxiliary_loss_mlp": 0.01026917,
      "balance_loss_clip": 1.01035416,
      "balance_loss_mlp": 1.01969838,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 1.9656801622116833,
      "language_loss": 0.78061795,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.80255145,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.756270170211792
    },
    {
      "auxiliary_loss_clip": 0.01151562,
      "auxiliary_loss_mlp": 0.01027081,
      "balance_loss_clip": 0.93113261,
      "balance_loss_mlp": 1.02030134,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 11.388086472271262,
      "language_loss": 0.69286788,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71465433,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 2.658958911895752
    },
    {
      "auxiliary_loss_clip": 0.01162605,
      "auxiliary_loss_mlp": 0.01022006,
      "balance_loss_clip": 0.85604966,
      "balance_loss_mlp": 1.01572359,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 1.8433228029783886,
      "language_loss": 0.70815742,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.73000348,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 3.581725835800171
    },
    {
      "auxiliary_loss_clip": 0.01165978,
      "auxiliary_loss_mlp": 0.01026845,
      "balance_loss_clip": 0.97368848,
      "balance_loss_mlp": 1.01998758,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 1.8230206083924914,
      "language_loss": 0.77909255,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.8010208,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.632222890853882
    },
    {
      "auxiliary_loss_clip": 0.01161301,
      "auxiliary_loss_mlp": 0.01031655,
      "balance_loss_clip": 0.81163675,
      "balance_loss_mlp": 1.02395427,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 2.9715121591738947,
      "language_loss": 0.68176317,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.70369267,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.8814780712127686
    },
    {
      "auxiliary_loss_clip": 0.01167349,
      "auxiliary_loss_mlp": 0.01029908,
      "balance_loss_clip": 0.97381127,
      "balance_loss_mlp": 1.02266026,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 1.8163513388174457,
      "language_loss": 0.71324003,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.73521256,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 2.8373050689697266
    },
    {
      "auxiliary_loss_clip": 0.01162439,
      "auxiliary_loss_mlp": 0.01031009,
      "balance_loss_clip": 0.89375389,
      "balance_loss_mlp": 1.02400589,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.8262505989129316,
      "language_loss": 0.7161057,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.73804015,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.643317937850952
    },
    {
      "auxiliary_loss_clip": 0.0116671,
      "auxiliary_loss_mlp": 0.01024539,
      "balance_loss_clip": 0.97067422,
      "balance_loss_mlp": 1.01730871,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.0266141305408056,
      "language_loss": 0.72030252,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74221492,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 3.6428062915802
    },
    {
      "auxiliary_loss_clip": 0.01164907,
      "auxiliary_loss_mlp": 0.01024537,
      "balance_loss_clip": 0.97009885,
      "balance_loss_mlp": 1.01728857,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.8764079852440294,
      "language_loss": 0.69958562,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.72148007,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.661123752593994
    },
    {
      "auxiliary_loss_clip": 0.01165473,
      "auxiliary_loss_mlp": 0.01026042,
      "balance_loss_clip": 1.01150417,
      "balance_loss_mlp": 1.01905322,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 1.7563273833974584,
      "language_loss": 0.84316874,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86508387,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 3.3823349475860596
    },
    {
      "auxiliary_loss_clip": 0.01163533,
      "auxiliary_loss_mlp": 0.0102363,
      "balance_loss_clip": 0.93287706,
      "balance_loss_mlp": 1.01646006,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.648850752980011,
      "language_loss": 0.7294879,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.75135946,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 3.740596055984497
    },
    {
      "auxiliary_loss_clip": 0.01153977,
      "auxiliary_loss_mlp": 0.01027133,
      "balance_loss_clip": 0.96926725,
      "balance_loss_mlp": 1.02048683,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 1.4459911554406082,
      "language_loss": 0.64818698,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.66999811,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 2.8156728744506836
    },
    {
      "auxiliary_loss_clip": 0.01165937,
      "auxiliary_loss_mlp": 0.01027061,
      "balance_loss_clip": 1.00895751,
      "balance_loss_mlp": 1.01919305,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.619353962087769,
      "language_loss": 0.72439975,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.74632978,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.6494545936584473
    },
    {
      "auxiliary_loss_clip": 0.0116768,
      "auxiliary_loss_mlp": 0.0102456,
      "balance_loss_clip": 1.0091114,
      "balance_loss_mlp": 1.01719332,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 3.0161229654128245,
      "language_loss": 0.63638598,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.65830839,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 2.6428706645965576
    },
    {
      "auxiliary_loss_clip": 0.01162872,
      "auxiliary_loss_mlp": 0.01020822,
      "balance_loss_clip": 1.00872087,
      "balance_loss_mlp": 1.0138247,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.765529953465917,
      "language_loss": 0.60310328,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.62494028,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 2.703974723815918
    },
    {
      "auxiliary_loss_clip": 0.01163782,
      "auxiliary_loss_mlp": 0.010269,
      "balance_loss_clip": 0.89946771,
      "balance_loss_mlp": 1.019557,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 1.9443127914223564,
      "language_loss": 0.76039869,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.78230548,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 2.7798025608062744
    },
    {
      "auxiliary_loss_clip": 0.01153188,
      "auxiliary_loss_mlp": 0.01122169,
      "balance_loss_clip": 1.00644076,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.6524909997404214,
      "language_loss": 0.87143874,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.89419228,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 2.648045778274536
    },
    {
      "auxiliary_loss_clip": 0.0116444,
      "auxiliary_loss_mlp": 0.01025075,
      "balance_loss_clip": 0.97147995,
      "balance_loss_mlp": 1.01828957,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 1.8406683536684376,
      "language_loss": 0.64696693,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.6688621,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 2.6390087604522705
    },
    {
      "auxiliary_loss_clip": 0.01161991,
      "auxiliary_loss_mlp": 0.01024881,
      "balance_loss_clip": 0.8960644,
      "balance_loss_mlp": 1.01793671,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 1.6241632344935508,
      "language_loss": 0.70996165,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73183036,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 2.7086427211761475
    },
    {
      "auxiliary_loss_clip": 0.01164564,
      "auxiliary_loss_mlp": 0.01020733,
      "balance_loss_clip": 1.04632187,
      "balance_loss_mlp": 1.01401889,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 2.29397293672201,
      "language_loss": 0.79656959,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.8184225,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 2.53609299659729
    },
    {
      "auxiliary_loss_clip": 0.01154834,
      "auxiliary_loss_mlp": 0.01025327,
      "balance_loss_clip": 1.00536704,
      "balance_loss_mlp": 1.01914275,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.5637351626784048,
      "language_loss": 0.76936555,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79116714,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 2.5988576412200928
    },
    {
      "auxiliary_loss_clip": 0.01153567,
      "auxiliary_loss_mlp": 0.01021917,
      "balance_loss_clip": 1.00555992,
      "balance_loss_mlp": 1.01544988,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 3.4671664120917405,
      "language_loss": 0.82755816,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.84931302,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.6148383617401123
    },
    {
      "auxiliary_loss_clip": 0.01160786,
      "auxiliary_loss_mlp": 0.01025636,
      "balance_loss_clip": 0.85575509,
      "balance_loss_mlp": 1.01846218,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 2.2245002505791627,
      "language_loss": 0.7435236,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76538777,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 2.831868886947632
    },
    {
      "auxiliary_loss_clip": 0.0116387,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 0.93154949,
      "balance_loss_mlp": 1.01878428,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 1.8250127851158722,
      "language_loss": 0.73738176,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.75927293,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 2.755688428878784
    },
    {
      "auxiliary_loss_clip": 0.01165201,
      "auxiliary_loss_mlp": 0.01019811,
      "balance_loss_clip": 1.0480895,
      "balance_loss_mlp": 1.01327264,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.796852686381444,
      "language_loss": 0.80897939,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.8308295,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.716134786605835
    },
    {
      "auxiliary_loss_clip": 0.01164021,
      "auxiliary_loss_mlp": 0.01032098,
      "balance_loss_clip": 1.00852931,
      "balance_loss_mlp": 1.02514553,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.7274559251834345,
      "language_loss": 0.7723068,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79426801,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.697411298751831
    },
    {
      "auxiliary_loss_clip": 0.01165297,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.04859376,
      "balance_loss_mlp": 1.0190115,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 1.682284702489171,
      "language_loss": 0.73448032,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.75639033,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.57793927192688
    },
    {
      "auxiliary_loss_clip": 0.01161288,
      "auxiliary_loss_mlp": 0.01021674,
      "balance_loss_clip": 0.89556056,
      "balance_loss_mlp": 1.01487947,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.7129398441276247,
      "language_loss": 0.82411575,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84594542,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.7264702320098877
    },
    {
      "auxiliary_loss_clip": 0.01155013,
      "auxiliary_loss_mlp": 0.0103024,
      "balance_loss_clip": 0.89084864,
      "balance_loss_mlp": 1.02327859,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 1.6698585650319209,
      "language_loss": 0.71718919,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.73904169,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 3.5435211658477783
    },
    {
      "auxiliary_loss_clip": 0.01156615,
      "auxiliary_loss_mlp": 0.01028843,
      "balance_loss_clip": 1.00656974,
      "balance_loss_mlp": 1.02157688,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 2.3063016083820203,
      "language_loss": 0.61102855,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63288313,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.6871187686920166
    },
    {
      "auxiliary_loss_clip": 0.01165856,
      "auxiliary_loss_mlp": 0.01024799,
      "balance_loss_clip": 1.04644084,
      "balance_loss_mlp": 1.01739597,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 1.9141785130980655,
      "language_loss": 0.84404469,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86595118,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.5636794567108154
    },
    {
      "auxiliary_loss_clip": 0.0116604,
      "auxiliary_loss_mlp": 0.01026058,
      "balance_loss_clip": 0.89380115,
      "balance_loss_mlp": 1.01872969,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.7254135350464064,
      "language_loss": 0.83456707,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.85648799,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.796412229537964
    },
    {
      "auxiliary_loss_clip": 0.01164245,
      "auxiliary_loss_mlp": 0.0102385,
      "balance_loss_clip": 0.97172534,
      "balance_loss_mlp": 1.01682258,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 2.1336377034638185,
      "language_loss": 0.8649475,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.88682848,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 2.6228413581848145
    },
    {
      "auxiliary_loss_clip": 0.01167174,
      "auxiliary_loss_mlp": 0.01121719,
      "balance_loss_clip": 1.01081622,
      "balance_loss_mlp": 0.0,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 2.062682496597517,
      "language_loss": 0.72627336,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.74916232,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.674788236618042
    },
    {
      "auxiliary_loss_clip": 0.011562,
      "auxiliary_loss_mlp": 0.01022927,
      "balance_loss_clip": 0.97042787,
      "balance_loss_mlp": 1.01621282,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.10635334561346,
      "language_loss": 0.86674958,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.88854086,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 4.063323736190796
    },
    {
      "auxiliary_loss_clip": 0.01166968,
      "auxiliary_loss_mlp": 0.010245,
      "balance_loss_clip": 1.04899025,
      "balance_loss_mlp": 1.01741278,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 1.837213850866575,
      "language_loss": 0.62750876,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.64942336,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 3.338597297668457
    },
    {
      "auxiliary_loss_clip": 0.01158128,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 0.96911305,
      "balance_loss_mlp": 1.01959229,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 1.940782302010755,
      "language_loss": 0.78803122,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.80987763,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 3.5027289390563965
    },
    {
      "auxiliary_loss_clip": 0.01165891,
      "auxiliary_loss_mlp": 0.01025422,
      "balance_loss_clip": 0.97577018,
      "balance_loss_mlp": 1.01847196,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.636783440766199,
      "language_loss": 0.75549573,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77740884,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 2.6282880306243896
    },
    {
      "auxiliary_loss_clip": 0.01171304,
      "auxiliary_loss_mlp": 0.01025398,
      "balance_loss_clip": 1.05004334,
      "balance_loss_mlp": 1.01844478,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 1.6869601365597904,
      "language_loss": 0.75611234,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77807933,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.5523829460144043
    },
    {
      "auxiliary_loss_clip": 0.01166184,
      "auxiliary_loss_mlp": 0.01026797,
      "balance_loss_clip": 0.93229771,
      "balance_loss_mlp": 1.01973736,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.2338459257035996,
      "language_loss": 0.69528455,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.71721435,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.6401474475860596
    },
    {
      "auxiliary_loss_clip": 0.01167634,
      "auxiliary_loss_mlp": 0.01026917,
      "balance_loss_clip": 1.04815495,
      "balance_loss_mlp": 1.01973438,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 2.37601757713046,
      "language_loss": 0.77101719,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79296267,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 2.5611658096313477
    },
    {
      "auxiliary_loss_clip": 0.01167369,
      "auxiliary_loss_mlp": 0.01025545,
      "balance_loss_clip": 1.04805899,
      "balance_loss_mlp": 1.0188694,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 2.3735185377230605,
      "language_loss": 0.7824223,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.80435145,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 2.545482635498047
    },
    {
      "auxiliary_loss_clip": 0.01155168,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 0.96608496,
      "balance_loss_mlp": 1.02094126,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 1.9395453029360152,
      "language_loss": 0.69669378,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.71852899,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 2.6766748428344727
    },
    {
      "auxiliary_loss_clip": 0.01149926,
      "auxiliary_loss_mlp": 0.0102687,
      "balance_loss_clip": 0.88770896,
      "balance_loss_mlp": 1.01911879,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 1.605147738779308,
      "language_loss": 0.73943472,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.76120269,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 2.7444136142730713
    },
    {
      "auxiliary_loss_clip": 0.01161618,
      "auxiliary_loss_mlp": 0.01025382,
      "balance_loss_clip": 0.96840239,
      "balance_loss_mlp": 1.01767516,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 2.0738705467260683,
      "language_loss": 0.7732774,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79514736,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 2.6619813442230225
    },
    {
      "auxiliary_loss_clip": 0.01161871,
      "auxiliary_loss_mlp": 0.01030293,
      "balance_loss_clip": 0.93166256,
      "balance_loss_mlp": 1.02280664,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 1.9297682776669784,
      "language_loss": 0.78801143,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.80993313,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.7442359924316406
    },
    {
      "auxiliary_loss_clip": 0.01159927,
      "auxiliary_loss_mlp": 0.01029132,
      "balance_loss_clip": 1.00762355,
      "balance_loss_mlp": 1.02194989,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 2.023337697795824,
      "language_loss": 0.62746692,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.6493575,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 2.6982197761535645
    },
    {
      "auxiliary_loss_clip": 0.01163873,
      "auxiliary_loss_mlp": 0.01020413,
      "balance_loss_clip": 0.97191638,
      "balance_loss_mlp": 1.01393688,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 2.0275823820423677,
      "language_loss": 0.85538101,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.87722385,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.671293020248413
    },
    {
      "auxiliary_loss_clip": 0.01166656,
      "auxiliary_loss_mlp": 0.01027181,
      "balance_loss_clip": 0.96952087,
      "balance_loss_mlp": 1.02037406,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 4.1249585642241655,
      "language_loss": 0.76014292,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78208131,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.6775224208831787
    },
    {
      "auxiliary_loss_clip": 0.01166735,
      "auxiliary_loss_mlp": 0.01027303,
      "balance_loss_clip": 0.97137594,
      "balance_loss_mlp": 1.02041268,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 2.054284530825727,
      "language_loss": 0.80384141,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82578182,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 2.659395217895508
    },
    {
      "auxiliary_loss_clip": 0.01170782,
      "auxiliary_loss_mlp": 0.01024246,
      "balance_loss_clip": 0.93262196,
      "balance_loss_mlp": 1.01737928,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 2.37608653682618,
      "language_loss": 0.83568293,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.85763317,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.6572704315185547
    },
    {
      "auxiliary_loss_clip": 0.01168252,
      "auxiliary_loss_mlp": 0.01025922,
      "balance_loss_clip": 1.00857997,
      "balance_loss_mlp": 1.01888847,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.577189538922842,
      "language_loss": 0.81267846,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83462024,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.6563608646392822
    },
    {
      "auxiliary_loss_clip": 0.01165785,
      "auxiliary_loss_mlp": 0.01028447,
      "balance_loss_clip": 0.93495685,
      "balance_loss_mlp": 1.02116299,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 2.5690640525901642,
      "language_loss": 0.81079704,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83273929,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.699491262435913
    },
    {
      "auxiliary_loss_clip": 0.01158516,
      "auxiliary_loss_mlp": 0.01021857,
      "balance_loss_clip": 0.96815777,
      "balance_loss_mlp": 1.01532769,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 1.726433242742184,
      "language_loss": 0.68965638,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.71146011,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 3.5092523097991943
    },
    {
      "auxiliary_loss_clip": 0.01170766,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.01350594,
      "balance_loss_mlp": 1.02038407,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 3.9871097096882604,
      "language_loss": 0.86765397,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.88964415,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.654512405395508
    },
    {
      "auxiliary_loss_clip": 0.01161448,
      "auxiliary_loss_mlp": 0.01026292,
      "balance_loss_clip": 0.89268512,
      "balance_loss_mlp": 1.01953316,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.5699430452875809,
      "language_loss": 0.76929724,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.79117465,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 2.76920485496521
    },
    {
      "auxiliary_loss_clip": 0.01166864,
      "auxiliary_loss_mlp": 0.01022322,
      "balance_loss_clip": 0.97014487,
      "balance_loss_mlp": 1.01570332,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 1.9902935942554114,
      "language_loss": 0.78653669,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.80842865,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.7458791732788086
    },
    {
      "auxiliary_loss_clip": 0.0115754,
      "auxiliary_loss_mlp": 0.01023202,
      "balance_loss_clip": 0.93159521,
      "balance_loss_mlp": 1.01669359,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 1.9400345546868816,
      "language_loss": 0.66583121,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.68763864,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.7005724906921387
    },
    {
      "auxiliary_loss_clip": 0.01163431,
      "auxiliary_loss_mlp": 0.01122092,
      "balance_loss_clip": 1.04700851,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 2.1507655944027375,
      "language_loss": 0.71557617,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.73843133,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 2.675295114517212
    },
    {
      "auxiliary_loss_clip": 0.01168146,
      "auxiliary_loss_mlp": 0.01023658,
      "balance_loss_clip": 0.8955934,
      "balance_loss_mlp": 1.01634741,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 3.721441303683059,
      "language_loss": 0.69581419,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71773219,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.7014925479888916
    },
    {
      "auxiliary_loss_clip": 0.01152752,
      "auxiliary_loss_mlp": 0.01122676,
      "balance_loss_clip": 0.92900622,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 1.961380295446067,
      "language_loss": 0.77594185,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.7986961,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 3.595404624938965
    },
    {
      "auxiliary_loss_clip": 0.01161604,
      "auxiliary_loss_mlp": 0.01021218,
      "balance_loss_clip": 0.96933198,
      "balance_loss_mlp": 1.01352954,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 1.9856932733106785,
      "language_loss": 0.62549984,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64732802,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 3.497791290283203
    },
    {
      "auxiliary_loss_clip": 0.01165155,
      "auxiliary_loss_mlp": 0.01023795,
      "balance_loss_clip": 1.0477773,
      "balance_loss_mlp": 1.01677406,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.5936991639227225,
      "language_loss": 0.61487931,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63676882,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 3.6075572967529297
    },
    {
      "auxiliary_loss_clip": 0.01159736,
      "auxiliary_loss_mlp": 0.01023519,
      "balance_loss_clip": 0.85545897,
      "balance_loss_mlp": 1.01717377,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.522931710889325,
      "language_loss": 0.78430092,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.80613345,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 2.8531715869903564
    },
    {
      "auxiliary_loss_clip": 0.01147302,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 0.92853391,
      "balance_loss_mlp": 1.02111125,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.7636797088202192,
      "language_loss": 0.83949071,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.86124408,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 3.0439765453338623
    },
    {
      "auxiliary_loss_clip": 0.01164658,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.0112772,
      "balance_loss_mlp": 1.0176369,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 1.5556469725212634,
      "language_loss": 0.77656221,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.79845232,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.591228723526001
    },
    {
      "auxiliary_loss_clip": 0.01173121,
      "auxiliary_loss_mlp": 0.01028111,
      "balance_loss_clip": 1.05073631,
      "balance_loss_mlp": 1.02094066,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 2.9147698938499524,
      "language_loss": 0.9122647,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.934277,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 2.6324830055236816
    },
    {
      "auxiliary_loss_clip": 0.01151024,
      "auxiliary_loss_mlp": 0.01025507,
      "balance_loss_clip": 0.89465225,
      "balance_loss_mlp": 1.01828003,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 2.360806784250362,
      "language_loss": 0.86650842,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.88827378,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 2.628408670425415
    },
    {
      "auxiliary_loss_clip": 0.01165319,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 1.0114429,
      "balance_loss_mlp": 1.02079189,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 2.2174826769485274,
      "language_loss": 0.80570161,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.82763499,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 2.568589687347412
    },
    {
      "auxiliary_loss_clip": 0.01167232,
      "auxiliary_loss_mlp": 0.01020962,
      "balance_loss_clip": 1.04772019,
      "balance_loss_mlp": 1.01433742,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 1.8513076881082235,
      "language_loss": 0.78275883,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80464077,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 2.5739781856536865
    },
    {
      "auxiliary_loss_clip": 0.01163644,
      "auxiliary_loss_mlp": 0.01023402,
      "balance_loss_clip": 0.97333056,
      "balance_loss_mlp": 1.01629734,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 1.6071001027192136,
      "language_loss": 0.7408874,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.7627579,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 2.6741175651550293
    },
    {
      "auxiliary_loss_clip": 0.01166183,
      "auxiliary_loss_mlp": 0.01024817,
      "balance_loss_clip": 1.0074122,
      "balance_loss_mlp": 1.01775336,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 2.1279783423148255,
      "language_loss": 0.66082698,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68273699,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 2.7875685691833496
    },
    {
      "auxiliary_loss_clip": 0.01170533,
      "auxiliary_loss_mlp": 0.0112284,
      "balance_loss_clip": 1.00906694,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 2.3480944243697475,
      "language_loss": 0.79125869,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81419241,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.7065694332122803
    },
    {
      "auxiliary_loss_clip": 0.01063631,
      "auxiliary_loss_mlp": 0.01000523,
      "balance_loss_clip": 0.97136682,
      "balance_loss_mlp": 0.9988777,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.9189921556234129,
      "language_loss": 0.5953353,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61597687,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.1572768688201904
    },
    {
      "auxiliary_loss_clip": 0.01163668,
      "auxiliary_loss_mlp": 0.01031731,
      "balance_loss_clip": 0.8166852,
      "balance_loss_mlp": 1.02438736,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 1.6308373266544136,
      "language_loss": 0.78771985,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.80967391,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 2.8733632564544678
    },
    {
      "auxiliary_loss_clip": 0.01166839,
      "auxiliary_loss_mlp": 0.0102697,
      "balance_loss_clip": 1.04855466,
      "balance_loss_mlp": 1.02004719,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 2.055947198029111,
      "language_loss": 0.65801632,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.67995441,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 3.1393933296203613
    },
    {
      "auxiliary_loss_clip": 0.01166342,
      "auxiliary_loss_mlp": 0.01122931,
      "balance_loss_clip": 0.8200165,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.147825011340169,
      "language_loss": 0.78234017,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.80523288,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.822631597518921
    },
    {
      "auxiliary_loss_clip": 0.0116267,
      "auxiliary_loss_mlp": 0.01025656,
      "balance_loss_clip": 0.97210062,
      "balance_loss_mlp": 1.01873589,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 1.8009385395131188,
      "language_loss": 0.74678469,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.76866794,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.731743097305298
    },
    {
      "auxiliary_loss_clip": 0.01156509,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.00518417,
      "balance_loss_mlp": 1.01986814,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 1.6796639268049212,
      "language_loss": 0.74606764,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.76790005,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.5870583057403564
    },
    {
      "auxiliary_loss_clip": 0.01168409,
      "auxiliary_loss_mlp": 0.01023827,
      "balance_loss_clip": 1.01101995,
      "balance_loss_mlp": 1.01657057,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 2.324750673770578,
      "language_loss": 0.81106174,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83298409,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 3.4971234798431396
    },
    {
      "auxiliary_loss_clip": 0.01157393,
      "auxiliary_loss_mlp": 0.01029045,
      "balance_loss_clip": 0.93320179,
      "balance_loss_mlp": 1.02229178,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 2.257475345502514,
      "language_loss": 0.81628555,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.83814996,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.7028493881225586
    },
    {
      "auxiliary_loss_clip": 0.01160115,
      "auxiliary_loss_mlp": 0.01026388,
      "balance_loss_clip": 1.00750089,
      "balance_loss_mlp": 1.01865137,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 1.961909374004028,
      "language_loss": 0.74211323,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.7639783,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 2.657979726791382
    },
    {
      "auxiliary_loss_clip": 0.01151072,
      "auxiliary_loss_mlp": 0.01030336,
      "balance_loss_clip": 0.88962722,
      "balance_loss_mlp": 1.02324331,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 2.4819059954949876,
      "language_loss": 0.76221538,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78402948,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.6929609775543213
    },
    {
      "auxiliary_loss_clip": 0.01170597,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 0.97077662,
      "balance_loss_mlp": 1.01828909,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 2.8439932462305078,
      "language_loss": 0.78587532,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.80783409,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.6246888637542725
    },
    {
      "auxiliary_loss_clip": 0.01164735,
      "auxiliary_loss_mlp": 0.01027525,
      "balance_loss_clip": 0.97192115,
      "balance_loss_mlp": 1.0202235,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 4.0348516554560305,
      "language_loss": 0.78376627,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.80568886,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.677550792694092
    },
    {
      "auxiliary_loss_clip": 0.01167077,
      "auxiliary_loss_mlp": 0.0102414,
      "balance_loss_clip": 1.04868484,
      "balance_loss_mlp": 1.01691878,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 2.325483185796491,
      "language_loss": 0.74277627,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.76468843,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 3.504950523376465
    },
    {
      "auxiliary_loss_clip": 0.01158067,
      "auxiliary_loss_mlp": 0.01022794,
      "balance_loss_clip": 1.00709188,
      "balance_loss_mlp": 1.01543319,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.1559832167733615,
      "language_loss": 0.74682075,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76862931,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.606785297393799
    },
    {
      "auxiliary_loss_clip": 0.01169848,
      "auxiliary_loss_mlp": 0.01026466,
      "balance_loss_clip": 0.97400045,
      "balance_loss_mlp": 1.01928413,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.821289388162289,
      "language_loss": 0.79652154,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81848466,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 3.6080760955810547
    },
    {
      "auxiliary_loss_clip": 0.01167687,
      "auxiliary_loss_mlp": 0.01023428,
      "balance_loss_clip": 1.04881811,
      "balance_loss_mlp": 1.01612055,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 2.5757618312991792,
      "language_loss": 0.81022811,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.83213925,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 3.5933752059936523
    },
    {
      "auxiliary_loss_clip": 0.0115692,
      "auxiliary_loss_mlp": 0.01025771,
      "balance_loss_clip": 0.93062955,
      "balance_loss_mlp": 1.018273,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 1.8827786495834762,
      "language_loss": 0.82804972,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.84987664,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.6592154502868652
    },
    {
      "auxiliary_loss_clip": 0.01062562,
      "auxiliary_loss_mlp": 0.0100105,
      "balance_loss_clip": 0.97071296,
      "balance_loss_mlp": 0.99945277,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7334822845007387,
      "language_loss": 0.61981881,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.64045489,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 3.3586838245391846
    },
    {
      "auxiliary_loss_clip": 0.01169124,
      "auxiliary_loss_mlp": 0.01025998,
      "balance_loss_clip": 0.93534315,
      "balance_loss_mlp": 1.01927149,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 1.750105531703065,
      "language_loss": 0.76517969,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78713089,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.6811821460723877
    },
    {
      "auxiliary_loss_clip": 0.01167951,
      "auxiliary_loss_mlp": 0.0102482,
      "balance_loss_clip": 0.85718334,
      "balance_loss_mlp": 1.01844537,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.589327021308972,
      "language_loss": 0.82436723,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.84629488,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 2.706904172897339
    },
    {
      "auxiliary_loss_clip": 0.01155285,
      "auxiliary_loss_mlp": 0.01020524,
      "balance_loss_clip": 1.00712776,
      "balance_loss_mlp": 1.01342762,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 1.6302998411094805,
      "language_loss": 0.62339407,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.64515215,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 2.6457910537719727
    },
    {
      "auxiliary_loss_clip": 0.01153647,
      "auxiliary_loss_mlp": 0.01022929,
      "balance_loss_clip": 0.88643688,
      "balance_loss_mlp": 1.01608086,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 3.5481680689319424,
      "language_loss": 0.7026397,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.72440541,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 2.8468244075775146
    },
    {
      "auxiliary_loss_clip": 0.01168601,
      "auxiliary_loss_mlp": 0.01028495,
      "balance_loss_clip": 1.00947726,
      "balance_loss_mlp": 1.02166688,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 2.0613762889478524,
      "language_loss": 0.78659999,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.80857092,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.6476645469665527
    },
    {
      "auxiliary_loss_clip": 0.01163276,
      "auxiliary_loss_mlp": 0.01122615,
      "balance_loss_clip": 0.97022796,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 1.8063733915043418,
      "language_loss": 0.69473505,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.71759391,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 2.783867597579956
    },
    {
      "auxiliary_loss_clip": 0.01156099,
      "auxiliary_loss_mlp": 0.01032733,
      "balance_loss_clip": 0.93176192,
      "balance_loss_mlp": 1.02524042,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.445153881924808,
      "language_loss": 0.61128402,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63317227,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 2.827836275100708
    },
    {
      "auxiliary_loss_clip": 0.01164111,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 0.89318514,
      "balance_loss_mlp": 1.02063727,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 1.8121218463116768,
      "language_loss": 0.76594901,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.78786564,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.775156259536743
    },
    {
      "auxiliary_loss_clip": 0.01164653,
      "auxiliary_loss_mlp": 0.01031518,
      "balance_loss_clip": 0.9312799,
      "balance_loss_mlp": 1.02405596,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 4.58812754347363,
      "language_loss": 0.88668501,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.9086467,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.6962828636169434
    },
    {
      "auxiliary_loss_clip": 0.01165937,
      "auxiliary_loss_mlp": 0.01025007,
      "balance_loss_clip": 0.9710651,
      "balance_loss_mlp": 1.01811719,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 2.018808515100643,
      "language_loss": 0.76685894,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.78876835,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 2.704176902770996
    },
    {
      "auxiliary_loss_clip": 0.01165013,
      "auxiliary_loss_mlp": 0.01023477,
      "balance_loss_clip": 1.00952005,
      "balance_loss_mlp": 1.01746309,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 1.8687219509552375,
      "language_loss": 0.79723388,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.8191188,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.7532074451446533
    },
    {
      "auxiliary_loss_clip": 0.01165262,
      "auxiliary_loss_mlp": 0.01027149,
      "balance_loss_clip": 1.04847157,
      "balance_loss_mlp": 1.02031541,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 1.9856363622143716,
      "language_loss": 0.81554461,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83746874,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.6372969150543213
    },
    {
      "auxiliary_loss_clip": 0.01175477,
      "auxiliary_loss_mlp": 0.01027114,
      "balance_loss_clip": 0.89461315,
      "balance_loss_mlp": 1.01910043,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 6.209481384776549,
      "language_loss": 0.70747483,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.72950077,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.813739538192749
    },
    {
      "auxiliary_loss_clip": 0.01166897,
      "auxiliary_loss_mlp": 0.01027869,
      "balance_loss_clip": 1.01028514,
      "balance_loss_mlp": 1.0213666,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 1.7373805790681054,
      "language_loss": 0.7066679,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.72861564,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.593583822250366
    },
    {
      "auxiliary_loss_clip": 0.01156926,
      "auxiliary_loss_mlp": 0.01026142,
      "balance_loss_clip": 1.00667453,
      "balance_loss_mlp": 1.01922154,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 1.589126895440396,
      "language_loss": 0.85460031,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87643099,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 3.450883388519287
    },
    {
      "auxiliary_loss_clip": 0.01164778,
      "auxiliary_loss_mlp": 0.01025844,
      "balance_loss_clip": 1.04470789,
      "balance_loss_mlp": 1.01902568,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 1.8642689487993624,
      "language_loss": 0.79103661,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.81294292,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.5658514499664307
    },
    {
      "auxiliary_loss_clip": 0.01161568,
      "auxiliary_loss_mlp": 0.01023277,
      "balance_loss_clip": 1.04588318,
      "balance_loss_mlp": 1.01644313,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.6167324517600334,
      "language_loss": 0.77018398,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79203248,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.6222074031829834
    },
    {
      "auxiliary_loss_clip": 0.0116551,
      "auxiliary_loss_mlp": 0.01028765,
      "balance_loss_clip": 1.00744617,
      "balance_loss_mlp": 1.02194619,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 1.9651519402281952,
      "language_loss": 0.75600457,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.77794731,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 2.6518003940582275
    },
    {
      "auxiliary_loss_clip": 0.01163286,
      "auxiliary_loss_mlp": 0.01023819,
      "balance_loss_clip": 0.97210789,
      "balance_loss_mlp": 1.01687169,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 1.6659928187684283,
      "language_loss": 0.80241263,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.82428366,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.7431020736694336
    },
    {
      "auxiliary_loss_clip": 0.01168111,
      "auxiliary_loss_mlp": 0.01026111,
      "balance_loss_clip": 1.0483098,
      "balance_loss_mlp": 1.01894021,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 2.5114690015520615,
      "language_loss": 0.82252097,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.84446311,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.548247814178467
    },
    {
      "auxiliary_loss_clip": 0.01159329,
      "auxiliary_loss_mlp": 0.01027532,
      "balance_loss_clip": 0.89087629,
      "balance_loss_mlp": 1.02052248,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 1.6115924560553503,
      "language_loss": 0.82575917,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.84762782,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 3.734238386154175
    },
    {
      "auxiliary_loss_clip": 0.01162795,
      "auxiliary_loss_mlp": 0.01024989,
      "balance_loss_clip": 0.969401,
      "balance_loss_mlp": 1.01804495,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 2.0110916823508975,
      "language_loss": 0.70900482,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.73088264,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.644770383834839
    },
    {
      "auxiliary_loss_clip": 0.01154992,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 0.89414632,
      "balance_loss_mlp": 1.0195725,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 1.9949097789183092,
      "language_loss": 0.82999521,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85180974,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 3.6339051723480225
    },
    {
      "auxiliary_loss_clip": 0.01167344,
      "auxiliary_loss_mlp": 0.01032559,
      "balance_loss_clip": 1.04795611,
      "balance_loss_mlp": 1.02551377,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 1.6880233597925276,
      "language_loss": 0.76446569,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.78646469,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.5885584354400635
    },
    {
      "auxiliary_loss_clip": 0.01158083,
      "auxiliary_loss_mlp": 0.01030592,
      "balance_loss_clip": 0.96986222,
      "balance_loss_mlp": 1.02280712,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 2.7045406279902027,
      "language_loss": 0.67312276,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.69500959,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 3.5385870933532715
    },
    {
      "auxiliary_loss_clip": 0.01164782,
      "auxiliary_loss_mlp": 0.01031607,
      "balance_loss_clip": 0.97200948,
      "balance_loss_mlp": 1.02426994,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 1.686730431732316,
      "language_loss": 0.7799536,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.80191749,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 2.635470151901245
    },
    {
      "auxiliary_loss_clip": 0.01159269,
      "auxiliary_loss_mlp": 0.01027106,
      "balance_loss_clip": 0.9670009,
      "balance_loss_mlp": 1.02045441,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.7458288968808906,
      "language_loss": 0.74509335,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.76695716,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.653898000717163
    },
    {
      "auxiliary_loss_clip": 0.01062264,
      "auxiliary_loss_mlp": 0.01000615,
      "balance_loss_clip": 0.97216856,
      "balance_loss_mlp": 0.99904138,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7628803729176076,
      "language_loss": 0.57027972,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.59090847,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.212756872177124
    },
    {
      "auxiliary_loss_clip": 0.01164018,
      "auxiliary_loss_mlp": 0.01027924,
      "balance_loss_clip": 0.93175912,
      "balance_loss_mlp": 1.0205394,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 3.3092557776795095,
      "language_loss": 0.68616205,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70808148,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 2.904552936553955
    },
    {
      "auxiliary_loss_clip": 0.01165464,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 0.93107665,
      "balance_loss_mlp": 1.01929116,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.6465876226137715,
      "language_loss": 0.79887903,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.82079262,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 2.691845178604126
    },
    {
      "auxiliary_loss_clip": 0.01164552,
      "auxiliary_loss_mlp": 0.01024514,
      "balance_loss_clip": 1.00944257,
      "balance_loss_mlp": 1.018291,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 1.7219340148677693,
      "language_loss": 0.72843552,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.75032616,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.602200508117676
    },
    {
      "auxiliary_loss_clip": 0.01168171,
      "auxiliary_loss_mlp": 0.01025804,
      "balance_loss_clip": 0.97077477,
      "balance_loss_mlp": 1.01868987,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 1.7433909235091205,
      "language_loss": 0.76160681,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.78354657,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 2.6817514896392822
    },
    {
      "auxiliary_loss_clip": 0.01145747,
      "auxiliary_loss_mlp": 0.01026585,
      "balance_loss_clip": 0.85238385,
      "balance_loss_mlp": 1.01937556,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 2.2331054936831762,
      "language_loss": 0.76432389,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.78604722,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 2.718035936355591
    },
    {
      "auxiliary_loss_clip": 0.01069306,
      "auxiliary_loss_mlp": 0.01001062,
      "balance_loss_clip": 0.90052658,
      "balance_loss_mlp": 0.99947661,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.7326930090826578,
      "language_loss": 0.55084932,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.57155299,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.2112441062927246
    },
    {
      "auxiliary_loss_clip": 0.01164991,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 0.97156233,
      "balance_loss_mlp": 1.02040327,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.362699240211662,
      "language_loss": 0.70492715,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.72685188,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.7075164318084717
    },
    {
      "auxiliary_loss_clip": 0.01060282,
      "auxiliary_loss_mlp": 0.01001507,
      "balance_loss_clip": 0.93287486,
      "balance_loss_mlp": 0.99977845,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8407150207237728,
      "language_loss": 0.62028605,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.64090395,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 3.2609753608703613
    },
    {
      "auxiliary_loss_clip": 0.01163926,
      "auxiliary_loss_mlp": 0.01024395,
      "balance_loss_clip": 1.00905943,
      "balance_loss_mlp": 1.01720691,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 1.9663172446103097,
      "language_loss": 0.76026994,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.78215319,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 2.7290422916412354
    },
    {
      "auxiliary_loss_clip": 0.01164561,
      "auxiliary_loss_mlp": 0.01027171,
      "balance_loss_clip": 1.04520082,
      "balance_loss_mlp": 1.01984596,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 2.505759809280163,
      "language_loss": 0.8286562,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.85057354,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.6376500129699707
    },
    {
      "auxiliary_loss_clip": 0.01161014,
      "auxiliary_loss_mlp": 0.01023252,
      "balance_loss_clip": 1.00902426,
      "balance_loss_mlp": 1.01656389,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 1.877775889113735,
      "language_loss": 0.70581728,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.72765994,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.6138811111450195
    },
    {
      "auxiliary_loss_clip": 0.01153949,
      "auxiliary_loss_mlp": 0.01027227,
      "balance_loss_clip": 0.81517196,
      "balance_loss_mlp": 1.02043164,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.0493081236473962,
      "language_loss": 0.80883014,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.83064187,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.86110258102417
    },
    {
      "auxiliary_loss_clip": 0.01165281,
      "auxiliary_loss_mlp": 0.01021385,
      "balance_loss_clip": 0.97170007,
      "balance_loss_mlp": 1.01444066,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 1.552906611045015,
      "language_loss": 0.75826395,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78013057,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 3.6357345581054688
    },
    {
      "auxiliary_loss_clip": 0.01169299,
      "auxiliary_loss_mlp": 0.01030719,
      "balance_loss_clip": 0.97423756,
      "balance_loss_mlp": 1.0237627,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 1.6978305661712025,
      "language_loss": 0.70493048,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.72693068,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.681908369064331
    },
    {
      "auxiliary_loss_clip": 0.01168993,
      "auxiliary_loss_mlp": 0.01028542,
      "balance_loss_clip": 1.01326561,
      "balance_loss_mlp": 1.02142859,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 1.8330021945219688,
      "language_loss": 0.77334476,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79532003,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 2.6829845905303955
    },
    {
      "auxiliary_loss_clip": 0.01172047,
      "auxiliary_loss_mlp": 0.01023132,
      "balance_loss_clip": 0.97423303,
      "balance_loss_mlp": 1.01642609,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 2.5244170639802697,
      "language_loss": 0.78459626,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.806548,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.730926275253296
    },
    {
      "auxiliary_loss_clip": 0.01163695,
      "auxiliary_loss_mlp": 0.01024335,
      "balance_loss_clip": 0.85445333,
      "balance_loss_mlp": 1.01705682,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 1.7617214241174923,
      "language_loss": 0.81805217,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.83993244,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.856912612915039
    },
    {
      "auxiliary_loss_clip": 0.01152173,
      "auxiliary_loss_mlp": 0.0102241,
      "balance_loss_clip": 0.93163669,
      "balance_loss_mlp": 1.01547837,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 1.571217008893138,
      "language_loss": 0.75263494,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.7743808,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.8724544048309326
    },
    {
      "auxiliary_loss_clip": 0.01168519,
      "auxiliary_loss_mlp": 0.01021667,
      "balance_loss_clip": 1.00690055,
      "balance_loss_mlp": 1.01435995,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.712406605602428,
      "language_loss": 0.78232551,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80422741,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 3.8984131813049316
    },
    {
      "auxiliary_loss_clip": 0.01172084,
      "auxiliary_loss_mlp": 0.01033508,
      "balance_loss_clip": 0.97195059,
      "balance_loss_mlp": 1.0259949,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 1.9660110315658659,
      "language_loss": 0.81689483,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83895075,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.6804897785186768
    },
    {
      "auxiliary_loss_clip": 0.01169716,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.01079774,
      "balance_loss_mlp": 1.01898301,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 1.9319036673637342,
      "language_loss": 0.69209117,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.7140584,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 3.7031545639038086
    },
    {
      "auxiliary_loss_clip": 0.01165188,
      "auxiliary_loss_mlp": 0.0102395,
      "balance_loss_clip": 1.00948238,
      "balance_loss_mlp": 1.01741195,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 1.9655176964086716,
      "language_loss": 0.72621548,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.7481069,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.631361961364746
    },
    {
      "auxiliary_loss_clip": 0.01172864,
      "auxiliary_loss_mlp": 0.01024505,
      "balance_loss_clip": 0.97238529,
      "balance_loss_mlp": 1.01741505,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 1.7713980684198756,
      "language_loss": 0.73436832,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.75634199,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 3.6951820850372314
    },
    {
      "auxiliary_loss_clip": 0.01165425,
      "auxiliary_loss_mlp": 0.01121965,
      "balance_loss_clip": 1.04789662,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 5.966715532263929,
      "language_loss": 0.78419745,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80707133,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.5641746520996094
    },
    {
      "auxiliary_loss_clip": 0.01168666,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04758334,
      "balance_loss_mlp": 1.01758933,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 1.8029659716642532,
      "language_loss": 0.76941407,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.79134619,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.6389894485473633
    },
    {
      "auxiliary_loss_clip": 0.01062477,
      "auxiliary_loss_mlp": 0.01001643,
      "balance_loss_clip": 0.93440735,
      "balance_loss_mlp": 1.0001049,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.7273379434810217,
      "language_loss": 0.65550792,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67614913,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 3.2602546215057373
    },
    {
      "auxiliary_loss_clip": 0.01169082,
      "auxiliary_loss_mlp": 0.01019631,
      "balance_loss_clip": 1.04835916,
      "balance_loss_mlp": 1.01288342,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 1.8956750826542534,
      "language_loss": 0.73617578,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.75806296,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 2.670753002166748
    },
    {
      "auxiliary_loss_clip": 0.01164839,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 1.04677773,
      "balance_loss_mlp": 1.020293,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 2.0506880056992465,
      "language_loss": 0.67173976,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.6936602,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 2.7649545669555664
    },
    {
      "auxiliary_loss_clip": 0.01167167,
      "auxiliary_loss_mlp": 0.01122575,
      "balance_loss_clip": 1.00823843,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 1.8775985644056856,
      "language_loss": 0.77592707,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.79882443,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.6302521228790283
    },
    {
      "auxiliary_loss_clip": 0.01169451,
      "auxiliary_loss_mlp": 0.01023782,
      "balance_loss_clip": 1.04946053,
      "balance_loss_mlp": 1.01642632,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.9312816003802395,
      "language_loss": 0.70184267,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72377503,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 2.623030662536621
    },
    {
      "auxiliary_loss_clip": 0.01168167,
      "auxiliary_loss_mlp": 0.01026814,
      "balance_loss_clip": 1.04674828,
      "balance_loss_mlp": 1.01969683,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 1.8203514287027966,
      "language_loss": 0.77570999,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79765975,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 2.6581764221191406
    },
    {
      "auxiliary_loss_clip": 0.01169686,
      "auxiliary_loss_mlp": 0.01020796,
      "balance_loss_clip": 1.04932666,
      "balance_loss_mlp": 1.0135448,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 3.274484149888707,
      "language_loss": 0.666897,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.68880177,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 2.645035743713379
    },
    {
      "auxiliary_loss_clip": 0.01169005,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 0.97130138,
      "balance_loss_mlp": 1.01895618,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.7106520148187574,
      "language_loss": 0.69005638,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.71200216,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.688229560852051
    },
    {
      "auxiliary_loss_clip": 0.01161403,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 0.96906793,
      "balance_loss_mlp": 1.02162349,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 2.0537132175701225,
      "language_loss": 0.78892595,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.81082416,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.666618824005127
    },
    {
      "auxiliary_loss_clip": 0.01062263,
      "auxiliary_loss_mlp": 0.01001903,
      "balance_loss_clip": 0.97042918,
      "balance_loss_mlp": 1.00013852,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 1.0044212892304918,
      "language_loss": 0.64910311,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66974473,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 3.1868321895599365
    },
    {
      "auxiliary_loss_clip": 0.01150991,
      "auxiliary_loss_mlp": 0.01023028,
      "balance_loss_clip": 0.96862739,
      "balance_loss_mlp": 1.01630187,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 1.9325350729555373,
      "language_loss": 0.74374771,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.76548791,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 2.721588611602783
    },
    {
      "auxiliary_loss_clip": 0.01159647,
      "auxiliary_loss_mlp": 0.01023958,
      "balance_loss_clip": 0.96913576,
      "balance_loss_mlp": 1.01654613,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.022807053661025,
      "language_loss": 0.65147537,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67331141,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.700669527053833
    },
    {
      "auxiliary_loss_clip": 0.0116399,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 0.93404758,
      "balance_loss_mlp": 1.02102816,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 1.891984371845231,
      "language_loss": 0.77989209,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.80181092,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.6865522861480713
    },
    {
      "auxiliary_loss_clip": 0.01171966,
      "auxiliary_loss_mlp": 0.01122803,
      "balance_loss_clip": 0.97152704,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 2.1582788023851336,
      "language_loss": 0.73803645,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.76098418,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.744878053665161
    },
    {
      "auxiliary_loss_clip": 0.01171607,
      "auxiliary_loss_mlp": 0.01030295,
      "balance_loss_clip": 1.00966907,
      "balance_loss_mlp": 1.0228591,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 2.1921253029188787,
      "language_loss": 0.66465485,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.68667388,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 3.3640637397766113
    },
    {
      "auxiliary_loss_clip": 0.011692,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.04817986,
      "balance_loss_mlp": 1.01824379,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 4.858283258891556,
      "language_loss": 0.6789335,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.70087945,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 2.6792633533477783
    },
    {
      "auxiliary_loss_clip": 0.01168292,
      "auxiliary_loss_mlp": 0.01028976,
      "balance_loss_clip": 1.04970622,
      "balance_loss_mlp": 1.02193379,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.5438303691099302,
      "language_loss": 0.73362601,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75559866,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.653642416000366
    },
    {
      "auxiliary_loss_clip": 0.01159929,
      "auxiliary_loss_mlp": 0.01023758,
      "balance_loss_clip": 0.9318127,
      "balance_loss_mlp": 1.0171417,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.0643506372303153,
      "language_loss": 0.74168611,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76352298,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.7449488639831543
    },
    {
      "auxiliary_loss_clip": 0.01163696,
      "auxiliary_loss_mlp": 0.01027101,
      "balance_loss_clip": 1.04585671,
      "balance_loss_mlp": 1.02009773,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 1.8167053861297857,
      "language_loss": 0.74496651,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.76687455,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.612234354019165
    },
    {
      "auxiliary_loss_clip": 0.01163474,
      "auxiliary_loss_mlp": 0.01031872,
      "balance_loss_clip": 0.89278865,
      "balance_loss_mlp": 1.02485681,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.608042058766756,
      "language_loss": 0.79353642,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.81548989,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 3.7222390174865723
    },
    {
      "auxiliary_loss_clip": 0.0117116,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.04898477,
      "balance_loss_mlp": 1.0264225,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 1.898228310766776,
      "language_loss": 0.78703928,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.80908835,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.633263349533081
    },
    {
      "auxiliary_loss_clip": 0.01166545,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.01219988,
      "balance_loss_mlp": 1.026636,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 1.938145476133956,
      "language_loss": 0.7160145,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.73801899,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.629352331161499
    },
    {
      "auxiliary_loss_clip": 0.01168282,
      "auxiliary_loss_mlp": 0.01023744,
      "balance_loss_clip": 1.04863667,
      "balance_loss_mlp": 1.0169462,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 2.0348446279259553,
      "language_loss": 0.77132118,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79324144,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 3.4781365394592285
    },
    {
      "auxiliary_loss_clip": 0.01162633,
      "auxiliary_loss_mlp": 0.01023021,
      "balance_loss_clip": 1.0090816,
      "balance_loss_mlp": 1.0162859,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 1.947747794509957,
      "language_loss": 0.74746341,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.76932001,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 3.5852794647216797
    },
    {
      "auxiliary_loss_clip": 0.0116573,
      "auxiliary_loss_mlp": 0.01021203,
      "balance_loss_clip": 0.9714421,
      "balance_loss_mlp": 1.01353204,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 1.6539227492695832,
      "language_loss": 0.65313327,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.67500263,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 2.748521327972412
    },
    {
      "auxiliary_loss_clip": 0.0116652,
      "auxiliary_loss_mlp": 0.01023789,
      "balance_loss_clip": 1.04784298,
      "balance_loss_mlp": 1.01705408,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.9629749174923372,
      "language_loss": 0.73653591,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.75843894,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.73752498626709
    },
    {
      "auxiliary_loss_clip": 0.01151844,
      "auxiliary_loss_mlp": 0.0102218,
      "balance_loss_clip": 0.89185643,
      "balance_loss_mlp": 1.01486373,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.7244587056431595,
      "language_loss": 0.73925906,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76099932,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 2.7799220085144043
    },
    {
      "auxiliary_loss_clip": 0.01170443,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.01178098,
      "balance_loss_mlp": 1.0190593,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.0830804797038818,
      "language_loss": 0.74225569,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76422727,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 2.7156453132629395
    },
    {
      "auxiliary_loss_clip": 0.01151227,
      "auxiliary_loss_mlp": 0.0101886,
      "balance_loss_clip": 0.92880988,
      "balance_loss_mlp": 1.01237798,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 2.142304210086046,
      "language_loss": 0.74020052,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.76190138,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 2.6548359394073486
    },
    {
      "auxiliary_loss_clip": 0.01167332,
      "auxiliary_loss_mlp": 0.0102293,
      "balance_loss_clip": 1.00803041,
      "balance_loss_mlp": 1.01635242,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.6262132093329835,
      "language_loss": 0.79267257,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.8145752,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.6352696418762207
    },
    {
      "auxiliary_loss_clip": 0.01164989,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 0.97088265,
      "balance_loss_mlp": 1.01718962,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 3.426823068048439,
      "language_loss": 0.76013875,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.78203559,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 2.5858941078186035
    },
    {
      "auxiliary_loss_clip": 0.010685,
      "auxiliary_loss_mlp": 0.01002398,
      "balance_loss_clip": 0.85986233,
      "balance_loss_mlp": 1.000705,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7918418660527632,
      "language_loss": 0.60820603,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.62891495,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.1730079650878906
    },
    {
      "auxiliary_loss_clip": 0.01164377,
      "auxiliary_loss_mlp": 0.0102707,
      "balance_loss_clip": 1.04841053,
      "balance_loss_mlp": 1.02031088,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 1.7645952310360746,
      "language_loss": 0.80175918,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.82367361,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.581282138824463
    },
    {
      "auxiliary_loss_clip": 0.01171731,
      "auxiliary_loss_mlp": 0.01024069,
      "balance_loss_clip": 0.93366784,
      "balance_loss_mlp": 1.01728582,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.7709863552538496,
      "language_loss": 0.76359451,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.78555256,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 2.7355403900146484
    },
    {
      "auxiliary_loss_clip": 0.01162635,
      "auxiliary_loss_mlp": 0.01122642,
      "balance_loss_clip": 0.89566755,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.5122292040812995,
      "language_loss": 0.7586627,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.78151548,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 2.7832400798797607
    },
    {
      "auxiliary_loss_clip": 0.0116294,
      "auxiliary_loss_mlp": 0.01027182,
      "balance_loss_clip": 0.97058761,
      "balance_loss_mlp": 1.02030349,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.6573774772829877,
      "language_loss": 0.83894223,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86084348,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 2.6429102420806885
    },
    {
      "auxiliary_loss_clip": 0.01163988,
      "auxiliary_loss_mlp": 0.01024278,
      "balance_loss_clip": 0.93133551,
      "balance_loss_mlp": 1.0174799,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 2.1562739710768017,
      "language_loss": 0.7271831,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.74906576,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.6259877681732178
    },
    {
      "auxiliary_loss_clip": 0.01163042,
      "auxiliary_loss_mlp": 0.01028468,
      "balance_loss_clip": 0.96979141,
      "balance_loss_mlp": 1.02145231,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 1.6958042106786102,
      "language_loss": 0.70145988,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72337496,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.713186740875244
    },
    {
      "auxiliary_loss_clip": 0.01169191,
      "auxiliary_loss_mlp": 0.01028383,
      "balance_loss_clip": 0.97116017,
      "balance_loss_mlp": 1.02151918,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.6527662051078698,
      "language_loss": 0.68753743,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.70951319,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.891246795654297
    },
    {
      "auxiliary_loss_clip": 0.01150418,
      "auxiliary_loss_mlp": 0.01023245,
      "balance_loss_clip": 0.92733175,
      "balance_loss_mlp": 1.01636696,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 1.6026836752598832,
      "language_loss": 0.84663904,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.86837572,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 3.572213649749756
    },
    {
      "auxiliary_loss_clip": 0.01167717,
      "auxiliary_loss_mlp": 0.01030216,
      "balance_loss_clip": 1.0102191,
      "balance_loss_mlp": 1.02339172,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 2.2336005712341263,
      "language_loss": 0.8231734,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84515274,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.6542553901672363
    },
    {
      "auxiliary_loss_clip": 0.01164589,
      "auxiliary_loss_mlp": 0.01025102,
      "balance_loss_clip": 1.00893986,
      "balance_loss_mlp": 1.01802707,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 1.579382220136877,
      "language_loss": 0.77592695,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79782391,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 2.68255615234375
    },
    {
      "auxiliary_loss_clip": 0.01151294,
      "auxiliary_loss_mlp": 0.01027208,
      "balance_loss_clip": 0.96786064,
      "balance_loss_mlp": 1.02027631,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 1.7532419490252398,
      "language_loss": 0.73961949,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.76140451,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.7071070671081543
    },
    {
      "auxiliary_loss_clip": 0.0116421,
      "auxiliary_loss_mlp": 0.01027491,
      "balance_loss_clip": 1.04666936,
      "balance_loss_mlp": 1.0205977,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 1.504312732017113,
      "language_loss": 0.76470417,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78662121,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 2.6435186862945557
    },
    {
      "auxiliary_loss_clip": 0.01155722,
      "auxiliary_loss_mlp": 0.01021758,
      "balance_loss_clip": 0.96902597,
      "balance_loss_mlp": 1.01444483,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.422925840657499,
      "language_loss": 0.64387029,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.66564506,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.660297393798828
    },
    {
      "auxiliary_loss_clip": 0.01168878,
      "auxiliary_loss_mlp": 0.01027674,
      "balance_loss_clip": 1.05155194,
      "balance_loss_mlp": 1.02085209,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 1.8813138537449308,
      "language_loss": 0.85146821,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87343377,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 3.5718588829040527
    },
    {
      "auxiliary_loss_clip": 0.01163928,
      "auxiliary_loss_mlp": 0.01122328,
      "balance_loss_clip": 1.00972128,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 1.9082494496561788,
      "language_loss": 0.8145383,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.83740085,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.689272880554199
    },
    {
      "auxiliary_loss_clip": 0.01167973,
      "auxiliary_loss_mlp": 0.0102466,
      "balance_loss_clip": 0.97628438,
      "balance_loss_mlp": 1.01768935,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 1.9724830490640735,
      "language_loss": 0.75260752,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77453387,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 3.5796611309051514
    },
    {
      "auxiliary_loss_clip": 0.01152299,
      "auxiliary_loss_mlp": 0.01028574,
      "balance_loss_clip": 0.93102348,
      "balance_loss_mlp": 1.02203834,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 1.7569672118032869,
      "language_loss": 0.70318949,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.72499824,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.7706892490386963
    },
    {
      "auxiliary_loss_clip": 0.01160114,
      "auxiliary_loss_mlp": 0.01026475,
      "balance_loss_clip": 0.97198665,
      "balance_loss_mlp": 1.0194118,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 1.5240072213371356,
      "language_loss": 0.71847713,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74034303,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 3.7434065341949463
    },
    {
      "auxiliary_loss_clip": 0.01169981,
      "auxiliary_loss_mlp": 0.01026924,
      "balance_loss_clip": 1.0122391,
      "balance_loss_mlp": 1.02041817,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.0111666047849988,
      "language_loss": 0.77185035,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79381943,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.611487627029419
    },
    {
      "auxiliary_loss_clip": 0.01167342,
      "auxiliary_loss_mlp": 0.01023977,
      "balance_loss_clip": 0.97038054,
      "balance_loss_mlp": 1.0169313,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 2.0961880324250424,
      "language_loss": 0.78968304,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.81159627,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.709087371826172
    },
    {
      "auxiliary_loss_clip": 0.01172038,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 0.93221664,
      "balance_loss_mlp": 1.02046812,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 1.9241377705338119,
      "language_loss": 0.79655576,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81855416,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 2.6777191162109375
    },
    {
      "auxiliary_loss_clip": 0.01162224,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 0.93515277,
      "balance_loss_mlp": 1.02036047,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.5498182434640115,
      "language_loss": 0.80041951,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.82232112,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.7629823684692383
    },
    {
      "auxiliary_loss_clip": 0.01164491,
      "auxiliary_loss_mlp": 0.0101983,
      "balance_loss_clip": 0.96846753,
      "balance_loss_mlp": 1.01299906,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.5923765064624076,
      "language_loss": 0.80664897,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.82849222,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 2.752312421798706
    },
    {
      "auxiliary_loss_clip": 0.01169025,
      "auxiliary_loss_mlp": 0.01027721,
      "balance_loss_clip": 1.04846597,
      "balance_loss_mlp": 1.02046967,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 2.6078968119886015,
      "language_loss": 0.7684558,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.79042327,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.756263017654419
    },
    {
      "auxiliary_loss_clip": 0.01058625,
      "auxiliary_loss_mlp": 0.01115537,
      "balance_loss_clip": 1.00773883,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8266263775524545,
      "language_loss": 0.63971663,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.66145825,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.2259814739227295
    },
    {
      "auxiliary_loss_clip": 0.01160125,
      "auxiliary_loss_mlp": 0.01025527,
      "balance_loss_clip": 0.89374483,
      "balance_loss_mlp": 1.01842785,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 1.828270014218134,
      "language_loss": 0.6998421,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72169858,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 2.728980779647827
    },
    {
      "auxiliary_loss_clip": 0.01162123,
      "auxiliary_loss_mlp": 0.01024103,
      "balance_loss_clip": 0.972565,
      "balance_loss_mlp": 1.01744843,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.5875210166620684,
      "language_loss": 0.64041084,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.66227311,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.6572353839874268
    },
    {
      "auxiliary_loss_clip": 0.01167118,
      "auxiliary_loss_mlp": 0.01122359,
      "balance_loss_clip": 1.04878676,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 3.1852769283661453,
      "language_loss": 0.74401414,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.76690888,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 2.579968214035034
    },
    {
      "auxiliary_loss_clip": 0.0116581,
      "auxiliary_loss_mlp": 0.0102453,
      "balance_loss_clip": 0.96716464,
      "balance_loss_mlp": 1.01703489,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 1.79728026564238,
      "language_loss": 0.7264505,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.74835384,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 2.812026023864746
    },
    {
      "auxiliary_loss_clip": 0.01157497,
      "auxiliary_loss_mlp": 0.01031251,
      "balance_loss_clip": 0.93557626,
      "balance_loss_mlp": 1.0240804,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.4879590130287903,
      "language_loss": 0.80688721,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.82877469,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.669309139251709
    },
    {
      "auxiliary_loss_clip": 0.01162912,
      "auxiliary_loss_mlp": 0.01021942,
      "balance_loss_clip": 0.92877054,
      "balance_loss_mlp": 1.01524234,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.806164666717163,
      "language_loss": 0.63622147,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.65807003,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.8545584678649902
    },
    {
      "auxiliary_loss_clip": 0.01159832,
      "auxiliary_loss_mlp": 0.01024637,
      "balance_loss_clip": 0.93365562,
      "balance_loss_mlp": 1.01720464,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 2.004598637347306,
      "language_loss": 0.70293105,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72477579,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.717125415802002
    },
    {
      "auxiliary_loss_clip": 0.01071955,
      "auxiliary_loss_mlp": 0.010035,
      "balance_loss_clip": 0.86194968,
      "balance_loss_mlp": 1.00186634,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.7409797552107715,
      "language_loss": 0.57238936,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59314388,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.282203197479248
    },
    {
      "auxiliary_loss_clip": 0.0105784,
      "auxiliary_loss_mlp": 0.01003582,
      "balance_loss_clip": 0.93535209,
      "balance_loss_mlp": 1.00210345,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.8008466240583862,
      "language_loss": 0.59006679,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.61068094,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.960994243621826
    },
    {
      "auxiliary_loss_clip": 0.01161766,
      "auxiliary_loss_mlp": 0.01024641,
      "balance_loss_clip": 0.9686594,
      "balance_loss_mlp": 1.01806021,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.5245818331902306,
      "language_loss": 0.77643871,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79830277,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.663984775543213
    },
    {
      "auxiliary_loss_clip": 0.01167385,
      "auxiliary_loss_mlp": 0.01020092,
      "balance_loss_clip": 0.97276711,
      "balance_loss_mlp": 1.01336563,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 1.7824264228835964,
      "language_loss": 0.76167858,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78355336,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 2.6597304344177246
    },
    {
      "auxiliary_loss_clip": 0.01170105,
      "auxiliary_loss_mlp": 0.01023829,
      "balance_loss_clip": 0.9695971,
      "balance_loss_mlp": 1.01645041,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.185135623756616,
      "language_loss": 0.6560607,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.67800009,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.7212259769439697
    },
    {
      "auxiliary_loss_clip": 0.01168946,
      "auxiliary_loss_mlp": 0.0102266,
      "balance_loss_clip": 1.04875469,
      "balance_loss_mlp": 1.01585579,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 1.909760507355269,
      "language_loss": 0.72705585,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.74897188,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 2.7532482147216797
    },
    {
      "auxiliary_loss_clip": 0.01168218,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 0.92976534,
      "balance_loss_mlp": 1.01610041,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 2.046033113248377,
      "language_loss": 0.74886537,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.77078521,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.787367820739746
    },
    {
      "auxiliary_loss_clip": 0.01170405,
      "auxiliary_loss_mlp": 0.01023538,
      "balance_loss_clip": 1.04913116,
      "balance_loss_mlp": 1.01710975,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 1.6567364055842952,
      "language_loss": 0.78694367,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.80888307,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 3.693730354309082
    },
    {
      "auxiliary_loss_clip": 0.01165655,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 0.93352973,
      "balance_loss_mlp": 1.02315021,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 1.8699346345468542,
      "language_loss": 0.76410902,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.78606391,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.6949989795684814
    },
    {
      "auxiliary_loss_clip": 0.01170428,
      "auxiliary_loss_mlp": 0.01021991,
      "balance_loss_clip": 1.01068461,
      "balance_loss_mlp": 1.01556814,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.716105981515384,
      "language_loss": 0.83558518,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.85750937,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.6156206130981445
    },
    {
      "auxiliary_loss_clip": 0.0116726,
      "auxiliary_loss_mlp": 0.01021908,
      "balance_loss_clip": 0.93199599,
      "balance_loss_mlp": 1.01563478,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 2.135495772600366,
      "language_loss": 0.79235083,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81424254,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 4.451690912246704
    },
    {
      "auxiliary_loss_clip": 0.01162785,
      "auxiliary_loss_mlp": 0.01025086,
      "balance_loss_clip": 0.92917007,
      "balance_loss_mlp": 1.01823735,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 2.3005055298909642,
      "language_loss": 0.6438235,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.66570228,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.7177956104278564
    },
    {
      "auxiliary_loss_clip": 0.01171375,
      "auxiliary_loss_mlp": 0.01026991,
      "balance_loss_clip": 0.97172701,
      "balance_loss_mlp": 1.02049136,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 1.7972573590971834,
      "language_loss": 0.80044162,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.82242525,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 2.618938446044922
    },
    {
      "auxiliary_loss_clip": 0.0116663,
      "auxiliary_loss_mlp": 0.01023674,
      "balance_loss_clip": 0.96871614,
      "balance_loss_mlp": 1.01729941,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 1.8159620108085566,
      "language_loss": 0.72876692,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.75066996,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 2.696841239929199
    },
    {
      "auxiliary_loss_clip": 0.01161558,
      "auxiliary_loss_mlp": 0.01024961,
      "balance_loss_clip": 0.97043473,
      "balance_loss_mlp": 1.01751018,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 3.1758200784520443,
      "language_loss": 0.87355161,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89541674,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 2.6698532104492188
    },
    {
      "auxiliary_loss_clip": 0.01164601,
      "auxiliary_loss_mlp": 0.01022435,
      "balance_loss_clip": 1.00679839,
      "balance_loss_mlp": 1.01531243,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.6567124978550032,
      "language_loss": 0.70291913,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.72478944,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 2.7082297801971436
    },
    {
      "auxiliary_loss_clip": 0.0115944,
      "auxiliary_loss_mlp": 0.01026111,
      "balance_loss_clip": 0.97136551,
      "balance_loss_mlp": 1.01959896,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 1.973323521564592,
      "language_loss": 0.78734583,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.80920136,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.656855821609497
    },
    {
      "auxiliary_loss_clip": 0.01158916,
      "auxiliary_loss_mlp": 0.01022768,
      "balance_loss_clip": 0.89263618,
      "balance_loss_mlp": 1.01606512,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 2.7458327823478226,
      "language_loss": 0.71394432,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.73576117,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 2.767383098602295
    },
    {
      "auxiliary_loss_clip": 0.01157629,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 0.93099469,
      "balance_loss_mlp": 1.02146983,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 2.1943051586867717,
      "language_loss": 0.64432114,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.66617787,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.873309850692749
    },
    {
      "auxiliary_loss_clip": 0.01163692,
      "auxiliary_loss_mlp": 0.01027471,
      "balance_loss_clip": 1.0122571,
      "balance_loss_mlp": 1.02070343,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 2.413375308805847,
      "language_loss": 0.82572114,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.84763277,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 2.7870311737060547
    },
    {
      "auxiliary_loss_clip": 0.01167938,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 0.89713323,
      "balance_loss_mlp": 1.02115083,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 2.242279544813859,
      "language_loss": 0.78912342,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.81108886,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 2.7142696380615234
    },
    {
      "auxiliary_loss_clip": 0.01169321,
      "auxiliary_loss_mlp": 0.01027776,
      "balance_loss_clip": 1.05039036,
      "balance_loss_mlp": 1.0205462,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.526109626823739,
      "language_loss": 0.71069896,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73266995,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 2.6970579624176025
    },
    {
      "auxiliary_loss_clip": 0.01163158,
      "auxiliary_loss_mlp": 0.01025653,
      "balance_loss_clip": 1.00813353,
      "balance_loss_mlp": 1.01863122,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 1.6836927162934174,
      "language_loss": 0.85292476,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87481284,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 2.6137430667877197
    },
    {
      "auxiliary_loss_clip": 0.01163887,
      "auxiliary_loss_mlp": 0.01123336,
      "balance_loss_clip": 1.00971913,
      "balance_loss_mlp": 0.0,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 2.9943148747684294,
      "language_loss": 0.68274844,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70562077,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 2.6255173683166504
    },
    {
      "auxiliary_loss_clip": 0.01160533,
      "auxiliary_loss_mlp": 0.01029861,
      "balance_loss_clip": 0.97091621,
      "balance_loss_mlp": 1.022017,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 1.839013325264406,
      "language_loss": 0.69860864,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.72051263,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.6707377433776855
    },
    {
      "auxiliary_loss_clip": 0.01163037,
      "auxiliary_loss_mlp": 0.01022369,
      "balance_loss_clip": 0.9721908,
      "balance_loss_mlp": 1.01552677,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 1.694359750613159,
      "language_loss": 0.78990602,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.81176007,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.5954716205596924
    },
    {
      "auxiliary_loss_clip": 0.01171324,
      "auxiliary_loss_mlp": 0.01122907,
      "balance_loss_clip": 0.97031599,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 4.829122125736299,
      "language_loss": 0.77767354,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.80061591,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.6486153602600098
    },
    {
      "auxiliary_loss_clip": 0.01162598,
      "auxiliary_loss_mlp": 0.01020649,
      "balance_loss_clip": 1.04522729,
      "balance_loss_mlp": 1.01389539,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 3.0662719353087926,
      "language_loss": 0.78622699,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80805945,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 3.320242404937744
    },
    {
      "auxiliary_loss_clip": 0.01167014,
      "auxiliary_loss_mlp": 0.01025659,
      "balance_loss_clip": 1.00954926,
      "balance_loss_mlp": 1.01764786,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 2.400829004620666,
      "language_loss": 0.7790516,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.8009783,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.6165757179260254
    },
    {
      "auxiliary_loss_clip": 0.01164073,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.00477481,
      "balance_loss_mlp": 1.01826823,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.1360964005713163,
      "language_loss": 0.73137248,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.7532692,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.5856869220733643
    },
    {
      "auxiliary_loss_clip": 0.01166996,
      "auxiliary_loss_mlp": 0.01023955,
      "balance_loss_clip": 1.04638553,
      "balance_loss_mlp": 1.01689816,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 1.6571100487530397,
      "language_loss": 0.64721429,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.66912383,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 3.025810956954956
    },
    {
      "auxiliary_loss_clip": 0.01168295,
      "auxiliary_loss_mlp": 0.01025964,
      "balance_loss_clip": 1.04957438,
      "balance_loss_mlp": 1.01879358,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 1.8837543619036201,
      "language_loss": 0.78377855,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.80572116,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.6156883239746094
    },
    {
      "auxiliary_loss_clip": 0.01165442,
      "auxiliary_loss_mlp": 0.01029161,
      "balance_loss_clip": 0.97121316,
      "balance_loss_mlp": 1.02256322,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 2.1067959931153837,
      "language_loss": 0.72352481,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74547088,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.648547887802124
    },
    {
      "auxiliary_loss_clip": 0.01159908,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 0.97197974,
      "balance_loss_mlp": 1.01682925,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.9282096386386702,
      "language_loss": 0.73666561,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.75850141,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 3.6107053756713867
    },
    {
      "auxiliary_loss_clip": 0.01164849,
      "auxiliary_loss_mlp": 0.01022693,
      "balance_loss_clip": 0.96930945,
      "balance_loss_mlp": 1.01588583,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.2379583432011287,
      "language_loss": 0.67407012,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69594556,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.6665444374084473
    },
    {
      "auxiliary_loss_clip": 0.01165819,
      "auxiliary_loss_mlp": 0.01021194,
      "balance_loss_clip": 1.04840815,
      "balance_loss_mlp": 1.01489127,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 1.9943840116426985,
      "language_loss": 0.74420112,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.7660712,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 3.562546968460083
    },
    {
      "auxiliary_loss_clip": 0.01166679,
      "auxiliary_loss_mlp": 0.01025933,
      "balance_loss_clip": 1.00948071,
      "balance_loss_mlp": 1.01935542,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 1.7101542308769377,
      "language_loss": 0.74201202,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76393813,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 3.6405811309814453
    },
    {
      "auxiliary_loss_clip": 0.01169002,
      "auxiliary_loss_mlp": 0.0102843,
      "balance_loss_clip": 1.00949585,
      "balance_loss_mlp": 1.02158141,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 2.0855433871947637,
      "language_loss": 0.73520386,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75717813,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 2.6625349521636963
    },
    {
      "auxiliary_loss_clip": 0.01168956,
      "auxiliary_loss_mlp": 0.01027774,
      "balance_loss_clip": 0.97324526,
      "balance_loss_mlp": 1.02062106,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 1.9378966038560415,
      "language_loss": 0.67298269,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69494998,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.7426161766052246
    },
    {
      "auxiliary_loss_clip": 0.01167587,
      "auxiliary_loss_mlp": 0.01122611,
      "balance_loss_clip": 0.97394878,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 2.8630546375021626,
      "language_loss": 0.85507697,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87797904,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.6757657527923584
    },
    {
      "auxiliary_loss_clip": 0.01167044,
      "auxiliary_loss_mlp": 0.01026755,
      "balance_loss_clip": 1.00888848,
      "balance_loss_mlp": 1.01925969,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 1.7937823719420452,
      "language_loss": 0.82119548,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.84313351,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 2.707486867904663
    },
    {
      "auxiliary_loss_clip": 0.01062313,
      "auxiliary_loss_mlp": 0.01115698,
      "balance_loss_clip": 0.97113591,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.7572630503230952,
      "language_loss": 0.63751364,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65929377,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 3.301461935043335
    },
    {
      "auxiliary_loss_clip": 0.01165573,
      "auxiliary_loss_mlp": 0.01025391,
      "balance_loss_clip": 1.00924134,
      "balance_loss_mlp": 1.01853609,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 4.120421731287104,
      "language_loss": 0.72106308,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.74297273,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.604741096496582
    },
    {
      "auxiliary_loss_clip": 0.01166569,
      "auxiliary_loss_mlp": 0.01021808,
      "balance_loss_clip": 1.00628805,
      "balance_loss_mlp": 1.01457465,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 3.702613060302514,
      "language_loss": 0.68813038,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.7100141,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 2.65435791015625
    },
    {
      "auxiliary_loss_clip": 0.01168879,
      "auxiliary_loss_mlp": 0.01026143,
      "balance_loss_clip": 1.01067412,
      "balance_loss_mlp": 1.01946723,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 1.481393553972857,
      "language_loss": 0.77624291,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.7981931,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.629558801651001
    },
    {
      "auxiliary_loss_clip": 0.01171111,
      "auxiliary_loss_mlp": 0.01030811,
      "balance_loss_clip": 0.93320906,
      "balance_loss_mlp": 1.02375114,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 1.9093345084788187,
      "language_loss": 0.81931818,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.84133744,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 2.697880506515503
    },
    {
      "auxiliary_loss_clip": 0.01167028,
      "auxiliary_loss_mlp": 0.01027267,
      "balance_loss_clip": 1.04648221,
      "balance_loss_mlp": 1.01979828,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 2.025338455533955,
      "language_loss": 0.79238337,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81432629,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.6505565643310547
    },
    {
      "auxiliary_loss_clip": 0.01168677,
      "auxiliary_loss_mlp": 0.01123063,
      "balance_loss_clip": 0.96999192,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 2.023404895728938,
      "language_loss": 0.62639284,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.64931023,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 2.663564920425415
    },
    {
      "auxiliary_loss_clip": 0.011561,
      "auxiliary_loss_mlp": 0.01020531,
      "balance_loss_clip": 0.92661983,
      "balance_loss_mlp": 1.01366723,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 1.929647192874403,
      "language_loss": 0.78566742,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80743372,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 2.751235246658325
    },
    {
      "auxiliary_loss_clip": 0.01169583,
      "auxiliary_loss_mlp": 0.01025941,
      "balance_loss_clip": 1.04916787,
      "balance_loss_mlp": 1.01920903,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 1.6330598678052755,
      "language_loss": 0.7031129,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.72506815,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 2.6528451442718506
    },
    {
      "auxiliary_loss_clip": 0.011666,
      "auxiliary_loss_mlp": 0.0112203,
      "balance_loss_clip": 1.01000285,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 3.245026689391919,
      "language_loss": 0.71297359,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.73585987,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.7165842056274414
    },
    {
      "auxiliary_loss_clip": 0.01167865,
      "auxiliary_loss_mlp": 0.01021448,
      "balance_loss_clip": 1.04794693,
      "balance_loss_mlp": 1.01430154,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 2.5534364088707044,
      "language_loss": 0.84219044,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86408353,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.568030834197998
    },
    {
      "auxiliary_loss_clip": 0.01163781,
      "auxiliary_loss_mlp": 0.01025578,
      "balance_loss_clip": 1.00844145,
      "balance_loss_mlp": 1.01887548,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 1.6987496750321902,
      "language_loss": 0.77022254,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.79211617,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.7390072345733643
    },
    {
      "auxiliary_loss_clip": 0.01164201,
      "auxiliary_loss_mlp": 0.0102377,
      "balance_loss_clip": 1.04807127,
      "balance_loss_mlp": 1.01703179,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 1.7482197778030912,
      "language_loss": 0.79866445,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.82054412,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 3.4092330932617188
    },
    {
      "auxiliary_loss_clip": 0.01170455,
      "auxiliary_loss_mlp": 0.01024232,
      "balance_loss_clip": 0.97215068,
      "balance_loss_mlp": 1.01709116,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 1.8178541523691594,
      "language_loss": 0.73803592,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.75998276,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.820831775665283
    },
    {
      "auxiliary_loss_clip": 0.01163087,
      "auxiliary_loss_mlp": 0.01027591,
      "balance_loss_clip": 1.0076524,
      "balance_loss_mlp": 1.02042103,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 1.874278176342814,
      "language_loss": 0.76320958,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.78511637,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 2.573911666870117
    },
    {
      "auxiliary_loss_clip": 0.01159268,
      "auxiliary_loss_mlp": 0.01020696,
      "balance_loss_clip": 1.00848699,
      "balance_loss_mlp": 1.01349592,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 2.563272729463867,
      "language_loss": 0.65927613,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.68107575,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 2.582169771194458
    },
    {
      "auxiliary_loss_clip": 0.0116899,
      "auxiliary_loss_mlp": 0.01023868,
      "balance_loss_clip": 1.00895071,
      "balance_loss_mlp": 1.01714766,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 1.689590281961893,
      "language_loss": 0.78216875,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.80409729,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.5888102054595947
    },
    {
      "auxiliary_loss_clip": 0.01159842,
      "auxiliary_loss_mlp": 0.01030159,
      "balance_loss_clip": 0.97366858,
      "balance_loss_mlp": 1.02339709,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.48179076793852,
      "language_loss": 0.78147984,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80337989,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.7053911685943604
    },
    {
      "auxiliary_loss_clip": 0.01170898,
      "auxiliary_loss_mlp": 0.01029546,
      "balance_loss_clip": 1.01075494,
      "balance_loss_mlp": 1.02239943,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.4317859547236926,
      "language_loss": 0.78623343,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.80823791,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 3.5508615970611572
    },
    {
      "auxiliary_loss_clip": 0.01157351,
      "auxiliary_loss_mlp": 0.01021228,
      "balance_loss_clip": 0.96978664,
      "balance_loss_mlp": 1.01413822,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 1.7864472294496618,
      "language_loss": 0.70741808,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.72920394,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.6759402751922607
    },
    {
      "auxiliary_loss_clip": 0.01164744,
      "auxiliary_loss_mlp": 0.01023386,
      "balance_loss_clip": 0.97077763,
      "balance_loss_mlp": 1.01628757,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 1.8143301576447144,
      "language_loss": 0.75094438,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77282566,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 3.614973545074463
    },
    {
      "auxiliary_loss_clip": 0.01158405,
      "auxiliary_loss_mlp": 0.01024252,
      "balance_loss_clip": 0.93394411,
      "balance_loss_mlp": 1.01757884,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 1.6501048388937838,
      "language_loss": 0.76883495,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.79066151,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 2.7723307609558105
    },
    {
      "auxiliary_loss_clip": 0.01170147,
      "auxiliary_loss_mlp": 0.01031013,
      "balance_loss_clip": 1.05080843,
      "balance_loss_mlp": 1.02349067,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 7.173717552307603,
      "language_loss": 0.81482399,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.8368355,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 3.5346248149871826
    },
    {
      "auxiliary_loss_clip": 0.01162213,
      "auxiliary_loss_mlp": 0.01028632,
      "balance_loss_clip": 0.96855932,
      "balance_loss_mlp": 1.02171755,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 2.2580338136000018,
      "language_loss": 0.77442944,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79633784,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.683598518371582
    },
    {
      "auxiliary_loss_clip": 0.01163607,
      "auxiliary_loss_mlp": 0.01024379,
      "balance_loss_clip": 0.97198486,
      "balance_loss_mlp": 1.01739979,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 1.824842908334017,
      "language_loss": 0.80933797,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.83121777,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.63075852394104
    },
    {
      "auxiliary_loss_clip": 0.01150154,
      "auxiliary_loss_mlp": 0.01023359,
      "balance_loss_clip": 0.92979228,
      "balance_loss_mlp": 1.01655507,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.7299605806655824,
      "language_loss": 0.78166234,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.80339754,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 2.699319839477539
    },
    {
      "auxiliary_loss_clip": 0.01164879,
      "auxiliary_loss_mlp": 0.01026885,
      "balance_loss_clip": 1.00873613,
      "balance_loss_mlp": 1.01984,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 2.2172905987668465,
      "language_loss": 0.73804921,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75996685,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 2.56595778465271
    },
    {
      "auxiliary_loss_clip": 0.01164914,
      "auxiliary_loss_mlp": 0.01027339,
      "balance_loss_clip": 1.04922903,
      "balance_loss_mlp": 1.02049017,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.7105705858765587,
      "language_loss": 0.75444925,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77637178,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.6203699111938477
    },
    {
      "auxiliary_loss_clip": 0.01150674,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 0.88934731,
      "balance_loss_mlp": 1.01838541,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 1.8286788235234186,
      "language_loss": 0.770311,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79207271,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 2.7095870971679688
    },
    {
      "auxiliary_loss_clip": 0.01166611,
      "auxiliary_loss_mlp": 0.01021485,
      "balance_loss_clip": 1.04568768,
      "balance_loss_mlp": 1.0144341,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 2.8104796804190335,
      "language_loss": 0.88312179,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90500277,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.608445882797241
    },
    {
      "auxiliary_loss_clip": 0.0106248,
      "auxiliary_loss_mlp": 0.01003098,
      "balance_loss_clip": 0.97030401,
      "balance_loss_mlp": 1.00150013,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7415613965064726,
      "language_loss": 0.6254316,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64608741,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 3.141789674758911
    },
    {
      "auxiliary_loss_clip": 0.01158269,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 0.97118151,
      "balance_loss_mlp": 1.0172987,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 1.9418094600586835,
      "language_loss": 0.675753,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.69757962,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.6425588130950928
    },
    {
      "auxiliary_loss_clip": 0.011633,
      "auxiliary_loss_mlp": 0.01025895,
      "balance_loss_clip": 1.01131475,
      "balance_loss_mlp": 1.01917744,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.18278598546064,
      "language_loss": 0.73700356,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.75889552,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 2.680410861968994
    },
    {
      "auxiliary_loss_clip": 0.01163435,
      "auxiliary_loss_mlp": 0.01022835,
      "balance_loss_clip": 1.00965333,
      "balance_loss_mlp": 1.01606429,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 1.7320181307790563,
      "language_loss": 0.78166652,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80352914,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 2.749485492706299
    },
    {
      "auxiliary_loss_clip": 0.0116358,
      "auxiliary_loss_mlp": 0.01026614,
      "balance_loss_clip": 0.93053836,
      "balance_loss_mlp": 1.01943433,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 1.7119090458218353,
      "language_loss": 0.74898022,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77088213,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 2.6876025199890137
    },
    {
      "auxiliary_loss_clip": 0.01152754,
      "auxiliary_loss_mlp": 0.010213,
      "balance_loss_clip": 0.9696902,
      "balance_loss_mlp": 1.01430488,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 2.54283853642527,
      "language_loss": 0.82622671,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.84796727,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.65634822845459
    },
    {
      "auxiliary_loss_clip": 0.01058741,
      "auxiliary_loss_mlp": 0.01002315,
      "balance_loss_clip": 1.00798774,
      "balance_loss_mlp": 1.00074148,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.8494932635027969,
      "language_loss": 0.58214259,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60275316,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.1690173149108887
    },
    {
      "auxiliary_loss_clip": 0.01150935,
      "auxiliary_loss_mlp": 0.01025841,
      "balance_loss_clip": 0.96727121,
      "balance_loss_mlp": 1.01889133,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.1043372251056214,
      "language_loss": 0.74822813,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76999587,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.6208813190460205
    },
    {
      "auxiliary_loss_clip": 0.01167075,
      "auxiliary_loss_mlp": 0.01122744,
      "balance_loss_clip": 1.00830758,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 1.7799596401075128,
      "language_loss": 0.76491362,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78781176,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 3.526820182800293
    },
    {
      "auxiliary_loss_clip": 0.01167068,
      "auxiliary_loss_mlp": 0.01028868,
      "balance_loss_clip": 1.04568422,
      "balance_loss_mlp": 1.02192998,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 1.8990979813177593,
      "language_loss": 0.71145165,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.73341095,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.6557562351226807
    },
    {
      "auxiliary_loss_clip": 0.01154046,
      "auxiliary_loss_mlp": 0.01030727,
      "balance_loss_clip": 0.93293929,
      "balance_loss_mlp": 1.0238874,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.5609460251076386,
      "language_loss": 0.80854017,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.83038795,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 2.781269073486328
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.01026037,
      "balance_loss_clip": 1.01027608,
      "balance_loss_mlp": 1.01927209,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 2.063786903351956,
      "language_loss": 0.69133008,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.71328187,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 2.823540449142456
    },
    {
      "auxiliary_loss_clip": 0.01164254,
      "auxiliary_loss_mlp": 0.01025446,
      "balance_loss_clip": 1.0095377,
      "balance_loss_mlp": 1.01820958,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 1.5502429894010983,
      "language_loss": 0.62295699,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.64485395,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.6842591762542725
    },
    {
      "auxiliary_loss_clip": 0.01162054,
      "auxiliary_loss_mlp": 0.01024355,
      "balance_loss_clip": 0.97179568,
      "balance_loss_mlp": 1.01758432,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 1.901343028587375,
      "language_loss": 0.61722374,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.63908786,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.6788594722747803
    },
    {
      "auxiliary_loss_clip": 0.01164812,
      "auxiliary_loss_mlp": 0.0102727,
      "balance_loss_clip": 0.97219723,
      "balance_loss_mlp": 1.02029371,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.2242367889722314,
      "language_loss": 0.53783536,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.55975616,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 3.712064027786255
    },
    {
      "auxiliary_loss_clip": 0.01163673,
      "auxiliary_loss_mlp": 0.01027759,
      "balance_loss_clip": 0.97163713,
      "balance_loss_mlp": 1.02086592,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 1.649056096534097,
      "language_loss": 0.83510727,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85702163,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.7373480796813965
    },
    {
      "auxiliary_loss_clip": 0.01171117,
      "auxiliary_loss_mlp": 0.01026842,
      "balance_loss_clip": 1.00879776,
      "balance_loss_mlp": 1.01930249,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 1.7378963421379603,
      "language_loss": 0.74857569,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.77055526,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.598417282104492
    },
    {
      "auxiliary_loss_clip": 0.01146804,
      "auxiliary_loss_mlp": 0.01026043,
      "balance_loss_clip": 0.81527245,
      "balance_loss_mlp": 1.01920605,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.866722613200292,
      "language_loss": 0.73395634,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75568485,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 3.681654930114746
    },
    {
      "auxiliary_loss_clip": 0.01163267,
      "auxiliary_loss_mlp": 0.01029601,
      "balance_loss_clip": 1.00990713,
      "balance_loss_mlp": 1.02305341,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 2.1546661826644313,
      "language_loss": 0.73923218,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76116085,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 3.608337163925171
    },
    {
      "auxiliary_loss_clip": 0.01153011,
      "auxiliary_loss_mlp": 0.01024959,
      "balance_loss_clip": 0.92983866,
      "balance_loss_mlp": 1.01755357,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 1.9229693745617058,
      "language_loss": 0.83443362,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.85621333,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.745122194290161
    },
    {
      "auxiliary_loss_clip": 0.01171343,
      "auxiliary_loss_mlp": 0.01027346,
      "balance_loss_clip": 1.00863004,
      "balance_loss_mlp": 1.01984823,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 1.8271470818525566,
      "language_loss": 0.7340405,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75602734,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.6689870357513428
    },
    {
      "auxiliary_loss_clip": 0.01163277,
      "auxiliary_loss_mlp": 0.01028205,
      "balance_loss_clip": 1.04740465,
      "balance_loss_mlp": 1.02103508,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 2.0285945021299945,
      "language_loss": 0.68075448,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.70266932,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 2.6325900554656982
    },
    {
      "auxiliary_loss_clip": 0.01165464,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 0.93245423,
      "balance_loss_mlp": 1.01820874,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.4730057497442366,
      "language_loss": 0.79714572,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.81904793,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 2.6994469165802
    },
    {
      "auxiliary_loss_clip": 0.01169614,
      "auxiliary_loss_mlp": 0.01122337,
      "balance_loss_clip": 1.00804877,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 2.4993629430274202,
      "language_loss": 0.7123946,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.73531407,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 2.6357109546661377
    },
    {
      "auxiliary_loss_clip": 0.01164353,
      "auxiliary_loss_mlp": 0.01026902,
      "balance_loss_clip": 0.96943802,
      "balance_loss_mlp": 1.01971388,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.5552298892981118,
      "language_loss": 0.84756958,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.8694821,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 2.6687307357788086
    },
    {
      "auxiliary_loss_clip": 0.01156862,
      "auxiliary_loss_mlp": 0.01027614,
      "balance_loss_clip": 0.96971029,
      "balance_loss_mlp": 1.01990104,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 2.7065607048201,
      "language_loss": 0.79518497,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81702971,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.715129852294922
    },
    {
      "auxiliary_loss_clip": 0.01162458,
      "auxiliary_loss_mlp": 0.01023903,
      "balance_loss_clip": 1.01001,
      "balance_loss_mlp": 1.01691711,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.2567945365641084,
      "language_loss": 0.79917186,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.8210355,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 2.70352840423584
    },
    {
      "auxiliary_loss_clip": 0.01164033,
      "auxiliary_loss_mlp": 0.01027209,
      "balance_loss_clip": 0.97109139,
      "balance_loss_mlp": 1.02015185,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 1.7091882591842418,
      "language_loss": 0.79163408,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.81354654,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 2.685194253921509
    },
    {
      "auxiliary_loss_clip": 0.0117225,
      "auxiliary_loss_mlp": 0.01021606,
      "balance_loss_clip": 0.97040164,
      "balance_loss_mlp": 1.01482916,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 1.69445443187095,
      "language_loss": 0.64897108,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67090964,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.64817476272583
    },
    {
      "auxiliary_loss_clip": 0.01163286,
      "auxiliary_loss_mlp": 0.01029847,
      "balance_loss_clip": 0.9361062,
      "balance_loss_mlp": 1.02317464,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 1.918667659547718,
      "language_loss": 0.86981714,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89174843,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 2.6997830867767334
    },
    {
      "auxiliary_loss_clip": 0.01166972,
      "auxiliary_loss_mlp": 0.01021527,
      "balance_loss_clip": 1.047369,
      "balance_loss_mlp": 1.01384366,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 1.9916033640408135,
      "language_loss": 0.75041902,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.77230406,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 2.548466920852661
    },
    {
      "auxiliary_loss_clip": 0.01163527,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.00853944,
      "balance_loss_mlp": 1.01995254,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.085083510167226,
      "language_loss": 0.78104711,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.80295199,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 2.636059045791626
    },
    {
      "auxiliary_loss_clip": 0.01066294,
      "auxiliary_loss_mlp": 0.01002761,
      "balance_loss_clip": 0.9334684,
      "balance_loss_mlp": 1.00109243,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7321858907032832,
      "language_loss": 0.56619668,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58688718,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.1120078563690186
    },
    {
      "auxiliary_loss_clip": 0.01160317,
      "auxiliary_loss_mlp": 0.01026038,
      "balance_loss_clip": 1.00886893,
      "balance_loss_mlp": 1.01871228,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 1.7372988190166059,
      "language_loss": 0.8688255,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.89068907,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.6006641387939453
    },
    {
      "auxiliary_loss_clip": 0.01170083,
      "auxiliary_loss_mlp": 0.0102831,
      "balance_loss_clip": 1.04672253,
      "balance_loss_mlp": 1.02041268,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 2.3769855270689915,
      "language_loss": 0.88492054,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.9069044,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 3.4335427284240723
    },
    {
      "auxiliary_loss_clip": 0.0117447,
      "auxiliary_loss_mlp": 0.01031779,
      "balance_loss_clip": 1.01289022,
      "balance_loss_mlp": 1.02377975,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 1.6818689572368506,
      "language_loss": 0.76655954,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.78862202,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.601515531539917
    },
    {
      "auxiliary_loss_clip": 0.0115668,
      "auxiliary_loss_mlp": 0.01028534,
      "balance_loss_clip": 0.89188087,
      "balance_loss_mlp": 1.02140808,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 1.8607880523218534,
      "language_loss": 0.74980062,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77165276,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.7305679321289062
    },
    {
      "auxiliary_loss_clip": 0.01163305,
      "auxiliary_loss_mlp": 0.01024138,
      "balance_loss_clip": 0.97269899,
      "balance_loss_mlp": 1.01710796,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 3.1392040983548424,
      "language_loss": 0.86650467,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.8883791,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 2.665193796157837
    },
    {
      "auxiliary_loss_clip": 0.01171335,
      "auxiliary_loss_mlp": 0.01122801,
      "balance_loss_clip": 1.01055348,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 1.4971156534197114,
      "language_loss": 0.67441607,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.69735742,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.745098114013672
    },
    {
      "auxiliary_loss_clip": 0.01148363,
      "auxiliary_loss_mlp": 0.0103101,
      "balance_loss_clip": 0.96561432,
      "balance_loss_mlp": 1.0244354,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 1.9648313248692308,
      "language_loss": 0.76225638,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.78405011,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.6194443702697754
    },
    {
      "auxiliary_loss_clip": 0.01168157,
      "auxiliary_loss_mlp": 0.01026701,
      "balance_loss_clip": 0.9714433,
      "balance_loss_mlp": 1.01972747,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.3175570625931794,
      "language_loss": 0.68974578,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.71169436,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 3.532392740249634
    },
    {
      "auxiliary_loss_clip": 0.0116346,
      "auxiliary_loss_mlp": 0.01023465,
      "balance_loss_clip": 0.93031132,
      "balance_loss_mlp": 1.01625872,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 2.188776800437161,
      "language_loss": 0.80243087,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.82430005,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.6352150440216064
    },
    {
      "auxiliary_loss_clip": 0.01171674,
      "auxiliary_loss_mlp": 0.01028808,
      "balance_loss_clip": 1.00790036,
      "balance_loss_mlp": 1.02147675,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 3.5896862233046223,
      "language_loss": 0.68473738,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70674217,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.6146035194396973
    },
    {
      "auxiliary_loss_clip": 0.01167474,
      "auxiliary_loss_mlp": 0.01030167,
      "balance_loss_clip": 0.86046875,
      "balance_loss_mlp": 1.0229671,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 2.18584540741048,
      "language_loss": 0.82274997,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.84472638,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 3.5930306911468506
    },
    {
      "auxiliary_loss_clip": 0.01165785,
      "auxiliary_loss_mlp": 0.01026665,
      "balance_loss_clip": 0.97035706,
      "balance_loss_mlp": 1.01973629,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 1.8614482371846568,
      "language_loss": 0.78132188,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80324638,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 3.424506425857544
    },
    {
      "auxiliary_loss_clip": 0.01161313,
      "auxiliary_loss_mlp": 0.01020815,
      "balance_loss_clip": 0.93107378,
      "balance_loss_mlp": 1.01397216,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 1.7573643456126673,
      "language_loss": 0.9203018,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94212306,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.7045767307281494
    },
    {
      "auxiliary_loss_clip": 0.01170591,
      "auxiliary_loss_mlp": 0.01122716,
      "balance_loss_clip": 1.01053798,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 3.1035592898640028,
      "language_loss": 0.78340739,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.80634046,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.6174421310424805
    },
    {
      "auxiliary_loss_clip": 0.01168669,
      "auxiliary_loss_mlp": 0.0103073,
      "balance_loss_clip": 1.04764247,
      "balance_loss_mlp": 1.02284408,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 1.8517406096340738,
      "language_loss": 0.78058422,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.80257827,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 2.616461992263794
    },
    {
      "auxiliary_loss_clip": 0.01170901,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.01161218,
      "balance_loss_mlp": 1.02369261,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.597070790262158,
      "language_loss": 0.82322407,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.84524691,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.7560555934906006
    },
    {
      "auxiliary_loss_clip": 0.011535,
      "auxiliary_loss_mlp": 0.01025431,
      "balance_loss_clip": 0.9691211,
      "balance_loss_mlp": 1.01792061,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.952155491127217,
      "language_loss": 0.77439725,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.79618657,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 2.670081377029419
    },
    {
      "auxiliary_loss_clip": 0.01168481,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.04744077,
      "balance_loss_mlp": 1.01926124,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 2.2832737791956794,
      "language_loss": 0.73944956,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.76139754,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 2.6295359134674072
    },
    {
      "auxiliary_loss_clip": 0.01161518,
      "auxiliary_loss_mlp": 0.01028809,
      "balance_loss_clip": 0.97229278,
      "balance_loss_mlp": 1.02186489,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 1.8575900518786608,
      "language_loss": 0.83525801,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.85716128,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.6460728645324707
    },
    {
      "auxiliary_loss_clip": 0.01153059,
      "auxiliary_loss_mlp": 0.01022868,
      "balance_loss_clip": 0.9653374,
      "balance_loss_mlp": 1.01580155,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.4897738714238438,
      "language_loss": 0.71141165,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73317087,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 2.714597463607788
    },
    {
      "auxiliary_loss_clip": 0.01161037,
      "auxiliary_loss_mlp": 0.01022596,
      "balance_loss_clip": 1.00730515,
      "balance_loss_mlp": 1.0155623,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 2.247156607069221,
      "language_loss": 0.82991594,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85175228,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.6249148845672607
    },
    {
      "auxiliary_loss_clip": 0.01165053,
      "auxiliary_loss_mlp": 0.01023821,
      "balance_loss_clip": 1.04653001,
      "balance_loss_mlp": 1.01662052,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 2.1270919612388766,
      "language_loss": 0.82668841,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84857714,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 2.5781993865966797
    },
    {
      "auxiliary_loss_clip": 0.01165373,
      "auxiliary_loss_mlp": 0.01026126,
      "balance_loss_clip": 1.04597449,
      "balance_loss_mlp": 1.01926589,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 10.178169987488825,
      "language_loss": 0.74887109,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.77078605,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 2.6384437084198
    },
    {
      "auxiliary_loss_clip": 0.01165108,
      "auxiliary_loss_mlp": 0.01022481,
      "balance_loss_clip": 1.04716313,
      "balance_loss_mlp": 1.01591277,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 2.278629047595554,
      "language_loss": 0.7864511,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80832696,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.648125648498535
    },
    {
      "auxiliary_loss_clip": 0.01172237,
      "auxiliary_loss_mlp": 0.01025034,
      "balance_loss_clip": 0.97049117,
      "balance_loss_mlp": 1.01675165,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 2.270823755415503,
      "language_loss": 0.69404393,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71601659,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 2.6171257495880127
    },
    {
      "auxiliary_loss_clip": 0.01166572,
      "auxiliary_loss_mlp": 0.01032882,
      "balance_loss_clip": 0.97132611,
      "balance_loss_mlp": 1.02487731,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.777823052522792,
      "language_loss": 0.84526384,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86725837,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 2.7098002433776855
    },
    {
      "auxiliary_loss_clip": 0.01165119,
      "auxiliary_loss_mlp": 0.01024912,
      "balance_loss_clip": 1.00993657,
      "balance_loss_mlp": 1.01785147,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 1.61499550243773,
      "language_loss": 0.70444483,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.72634518,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.653416156768799
    },
    {
      "auxiliary_loss_clip": 0.01167352,
      "auxiliary_loss_mlp": 0.01027427,
      "balance_loss_clip": 0.93478996,
      "balance_loss_mlp": 1.01956546,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 1.7827331413802319,
      "language_loss": 0.74812347,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.77007127,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 3.603902816772461
    },
    {
      "auxiliary_loss_clip": 0.01162853,
      "auxiliary_loss_mlp": 0.0102234,
      "balance_loss_clip": 0.93355876,
      "balance_loss_mlp": 1.01537204,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.2031046196323567,
      "language_loss": 0.88511872,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.90697062,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.7465591430664062
    },
    {
      "auxiliary_loss_clip": 0.01169488,
      "auxiliary_loss_mlp": 0.01122649,
      "balance_loss_clip": 0.89784831,
      "balance_loss_mlp": 0.0,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 1.5577576317856123,
      "language_loss": 0.72578382,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.74870521,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 2.708718776702881
    },
    {
      "auxiliary_loss_clip": 0.01155583,
      "auxiliary_loss_mlp": 0.01025601,
      "balance_loss_clip": 0.96994478,
      "balance_loss_mlp": 1.01892281,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 1.4901203186762906,
      "language_loss": 0.7073639,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.72917581,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 2.6301419734954834
    },
    {
      "auxiliary_loss_clip": 0.01170141,
      "auxiliary_loss_mlp": 0.01121876,
      "balance_loss_clip": 1.01002026,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 1.7879019115321642,
      "language_loss": 0.84078157,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.8637017,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.6054420471191406
    },
    {
      "auxiliary_loss_clip": 0.01160366,
      "auxiliary_loss_mlp": 0.01019814,
      "balance_loss_clip": 0.96837068,
      "balance_loss_mlp": 1.01298356,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 1.7708341011907471,
      "language_loss": 0.79230142,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.81410325,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 3.575986862182617
    },
    {
      "auxiliary_loss_clip": 0.0115989,
      "auxiliary_loss_mlp": 0.0103243,
      "balance_loss_clip": 0.92857254,
      "balance_loss_mlp": 1.02497983,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 2.0810218148463937,
      "language_loss": 0.67751586,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.69943899,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.7038466930389404
    },
    {
      "auxiliary_loss_clip": 0.01158461,
      "auxiliary_loss_mlp": 0.0102252,
      "balance_loss_clip": 0.96981168,
      "balance_loss_mlp": 1.01599312,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 1.9884668622271795,
      "language_loss": 0.79242527,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.81423509,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.612330913543701
    },
    {
      "auxiliary_loss_clip": 0.01163658,
      "auxiliary_loss_mlp": 0.01024228,
      "balance_loss_clip": 1.00851536,
      "balance_loss_mlp": 1.01693821,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 1.9085927935660927,
      "language_loss": 0.75372213,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.77560097,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 3.522387981414795
    },
    {
      "auxiliary_loss_clip": 0.01167147,
      "auxiliary_loss_mlp": 0.01027113,
      "balance_loss_clip": 0.97285771,
      "balance_loss_mlp": 1.01988947,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 1.6225809575815189,
      "language_loss": 0.73577082,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.75771344,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.6042582988739014
    },
    {
      "auxiliary_loss_clip": 0.01164539,
      "auxiliary_loss_mlp": 0.01027821,
      "balance_loss_clip": 0.97168291,
      "balance_loss_mlp": 1.02074254,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 1.6807398672539027,
      "language_loss": 0.68364942,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70557296,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 3.6563565731048584
    },
    {
      "auxiliary_loss_clip": 0.01167567,
      "auxiliary_loss_mlp": 0.01023653,
      "balance_loss_clip": 1.04879642,
      "balance_loss_mlp": 1.0168432,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 2.704819536065786,
      "language_loss": 0.71608627,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.73799849,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.60947322845459
    },
    {
      "auxiliary_loss_clip": 0.01157205,
      "auxiliary_loss_mlp": 0.0102154,
      "balance_loss_clip": 1.00696611,
      "balance_loss_mlp": 1.01487064,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.0234260155167134,
      "language_loss": 0.84958065,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87136805,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.5784993171691895
    },
    {
      "auxiliary_loss_clip": 0.01063389,
      "auxiliary_loss_mlp": 0.0100116,
      "balance_loss_clip": 0.97107029,
      "balance_loss_mlp": 0.99949139,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.7564703817760566,
      "language_loss": 0.59676564,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61741114,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 3.2861075401306152
    },
    {
      "auxiliary_loss_clip": 0.01164062,
      "auxiliary_loss_mlp": 0.01025884,
      "balance_loss_clip": 0.73982495,
      "balance_loss_mlp": 1.01842189,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 1.7130828504536875,
      "language_loss": 0.57189339,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59379292,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 3.1485161781311035
    },
    {
      "auxiliary_loss_clip": 0.01065344,
      "auxiliary_loss_mlp": 0.01001063,
      "balance_loss_clip": 0.93404418,
      "balance_loss_mlp": 0.9994179,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.7957032020148154,
      "language_loss": 0.57797241,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59863651,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.4020612239837646
    },
    {
      "auxiliary_loss_clip": 0.01163048,
      "auxiliary_loss_mlp": 0.01019926,
      "balance_loss_clip": 0.97169292,
      "balance_loss_mlp": 1.01302135,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 1.8431602599147534,
      "language_loss": 0.74960583,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.77143556,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 2.669233560562134
    },
    {
      "auxiliary_loss_clip": 0.01171942,
      "auxiliary_loss_mlp": 0.01023229,
      "balance_loss_clip": 0.93242675,
      "balance_loss_mlp": 1.01621628,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 1.911235950663936,
      "language_loss": 0.58028054,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.60223222,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.7503201961517334
    },
    {
      "auxiliary_loss_clip": 0.0116644,
      "auxiliary_loss_mlp": 0.01028113,
      "balance_loss_clip": 0.97200722,
      "balance_loss_mlp": 1.02110636,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 2.005528337030787,
      "language_loss": 0.68096292,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70290852,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 2.6831257343292236
    },
    {
      "auxiliary_loss_clip": 0.01153762,
      "auxiliary_loss_mlp": 0.01023031,
      "balance_loss_clip": 1.0051105,
      "balance_loss_mlp": 1.01658225,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 1.6885799276890059,
      "language_loss": 0.81635046,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.83811843,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.6012227535247803
    },
    {
      "auxiliary_loss_clip": 0.01161896,
      "auxiliary_loss_mlp": 0.01022833,
      "balance_loss_clip": 0.97023892,
      "balance_loss_mlp": 1.01546037,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 1.681301787940304,
      "language_loss": 0.81844187,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.84028912,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.706312894821167
    },
    {
      "auxiliary_loss_clip": 0.01161513,
      "auxiliary_loss_mlp": 0.01028575,
      "balance_loss_clip": 0.85551476,
      "balance_loss_mlp": 1.02124941,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 2.2270588101626294,
      "language_loss": 0.67015737,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.69205827,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 2.7101423740386963
    },
    {
      "auxiliary_loss_clip": 0.0116563,
      "auxiliary_loss_mlp": 0.01029268,
      "balance_loss_clip": 0.97187614,
      "balance_loss_mlp": 1.02186489,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 1.8764539824710533,
      "language_loss": 0.72606409,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.74801302,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.7864668369293213
    },
    {
      "auxiliary_loss_clip": 0.01057972,
      "auxiliary_loss_mlp": 0.01004436,
      "balance_loss_clip": 0.97068322,
      "balance_loss_mlp": 1.00282657,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6884892867539464,
      "language_loss": 0.55722958,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57785362,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 3.244715452194214
    },
    {
      "auxiliary_loss_clip": 0.01163878,
      "auxiliary_loss_mlp": 0.01023095,
      "balance_loss_clip": 0.97245914,
      "balance_loss_mlp": 1.01628494,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 3.315161488012721,
      "language_loss": 0.75865501,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.78052473,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.64937162399292
    },
    {
      "auxiliary_loss_clip": 0.01155953,
      "auxiliary_loss_mlp": 0.01122945,
      "balance_loss_clip": 0.89071596,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 1.8954823982252829,
      "language_loss": 0.74850261,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.77129155,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.718994140625
    },
    {
      "auxiliary_loss_clip": 0.01163975,
      "auxiliary_loss_mlp": 0.01122061,
      "balance_loss_clip": 1.01218998,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 2.0762132268741293,
      "language_loss": 0.74586105,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.7687214,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 3.687988758087158
    },
    {
      "auxiliary_loss_clip": 0.01169413,
      "auxiliary_loss_mlp": 0.01029139,
      "balance_loss_clip": 0.85679269,
      "balance_loss_mlp": 1.02149189,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 2.228006070031152,
      "language_loss": 0.79290116,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.81488663,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 2.7914583683013916
    },
    {
      "auxiliary_loss_clip": 0.0116972,
      "auxiliary_loss_mlp": 0.01030747,
      "balance_loss_clip": 1.01058888,
      "balance_loss_mlp": 1.02383304,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.9372127172430165,
      "language_loss": 0.74339521,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.76539981,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 2.714470863342285
    },
    {
      "auxiliary_loss_clip": 0.01168944,
      "auxiliary_loss_mlp": 0.01025026,
      "balance_loss_clip": 1.01228142,
      "balance_loss_mlp": 1.01762319,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 5.0425449297855565,
      "language_loss": 0.77690071,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.7988404,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.7028310298919678
    },
    {
      "auxiliary_loss_clip": 0.01166586,
      "auxiliary_loss_mlp": 0.01023613,
      "balance_loss_clip": 1.04761434,
      "balance_loss_mlp": 1.01627517,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 1.9093809129185608,
      "language_loss": 0.76250553,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.7844075,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.5662460327148438
    },
    {
      "auxiliary_loss_clip": 0.01158855,
      "auxiliary_loss_mlp": 0.01031082,
      "balance_loss_clip": 0.89766383,
      "balance_loss_mlp": 1.02384043,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 1.9885576995013783,
      "language_loss": 0.67458075,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.69648015,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 3.774634838104248
    },
    {
      "auxiliary_loss_clip": 0.01161108,
      "auxiliary_loss_mlp": 0.01025133,
      "balance_loss_clip": 1.00827622,
      "balance_loss_mlp": 1.01774228,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 1.8351920369774257,
      "language_loss": 0.69112778,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.71299016,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.599045991897583
    },
    {
      "auxiliary_loss_clip": 0.01059568,
      "auxiliary_loss_mlp": 0.01003573,
      "balance_loss_clip": 0.97302204,
      "balance_loss_mlp": 1.00198746,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.9332537443443354,
      "language_loss": 0.61181843,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.63244987,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 3.2263739109039307
    },
    {
      "auxiliary_loss_clip": 0.01164087,
      "auxiliary_loss_mlp": 0.01029657,
      "balance_loss_clip": 1.04655552,
      "balance_loss_mlp": 1.02270985,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 2.6806733896434114,
      "language_loss": 0.83354062,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85547805,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 3.1562273502349854
    },
    {
      "auxiliary_loss_clip": 0.01160981,
      "auxiliary_loss_mlp": 0.01028874,
      "balance_loss_clip": 0.9301194,
      "balance_loss_mlp": 1.02150679,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 2.400264006886677,
      "language_loss": 0.80939674,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.83129525,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.590139389038086
    },
    {
      "auxiliary_loss_clip": 0.01167325,
      "auxiliary_loss_mlp": 0.01024117,
      "balance_loss_clip": 1.01016951,
      "balance_loss_mlp": 1.01713705,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 2.044912976286535,
      "language_loss": 0.85865539,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.88056982,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 3.6871304512023926
    },
    {
      "auxiliary_loss_clip": 0.0115719,
      "auxiliary_loss_mlp": 0.01031233,
      "balance_loss_clip": 0.96776879,
      "balance_loss_mlp": 1.02390182,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 1.9778840746965245,
      "language_loss": 0.82568908,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.8475734,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.7049789428710938
    },
    {
      "auxiliary_loss_clip": 0.01168235,
      "auxiliary_loss_mlp": 0.01029033,
      "balance_loss_clip": 0.97086477,
      "balance_loss_mlp": 1.02193403,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 7.285446979034458,
      "language_loss": 0.83219236,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85416496,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.65335750579834
    },
    {
      "auxiliary_loss_clip": 0.01171396,
      "auxiliary_loss_mlp": 0.01122554,
      "balance_loss_clip": 0.97117901,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 1.9685696773811991,
      "language_loss": 0.81052542,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.83346486,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 2.6198856830596924
    },
    {
      "auxiliary_loss_clip": 0.01171879,
      "auxiliary_loss_mlp": 0.01029259,
      "balance_loss_clip": 1.00959897,
      "balance_loss_mlp": 1.02149558,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 2.27458559067366,
      "language_loss": 0.70926082,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73127222,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.541604518890381
    },
    {
      "auxiliary_loss_clip": 0.01173208,
      "auxiliary_loss_mlp": 0.01026789,
      "balance_loss_clip": 1.05087948,
      "balance_loss_mlp": 1.01887393,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 2.3944288405891703,
      "language_loss": 0.61391163,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63591164,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.4770357608795166
    },
    {
      "auxiliary_loss_clip": 0.01166234,
      "auxiliary_loss_mlp": 0.01027715,
      "balance_loss_clip": 1.0475173,
      "balance_loss_mlp": 1.02113152,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 8.0034490302738,
      "language_loss": 0.55648458,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.5784241,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 2.473422050476074
    },
    {
      "auxiliary_loss_clip": 0.01172228,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.01023829,
      "balance_loss_mlp": 1.02045512,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 10.24225867979148,
      "language_loss": 0.79994702,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.82194883,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.588190793991089
    },
    {
      "auxiliary_loss_clip": 0.01164088,
      "auxiliary_loss_mlp": 0.01122874,
      "balance_loss_clip": 0.97204781,
      "balance_loss_mlp": 0.0,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 2.0658297320219945,
      "language_loss": 0.63019931,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.65306902,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 2.7014319896698
    },
    {
      "auxiliary_loss_clip": 0.01172304,
      "auxiliary_loss_mlp": 0.01030418,
      "balance_loss_clip": 0.85686976,
      "balance_loss_mlp": 1.02282393,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 2.44105861741753,
      "language_loss": 0.85905492,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.88108218,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.7684152126312256
    },
    {
      "auxiliary_loss_clip": 0.01159463,
      "auxiliary_loss_mlp": 0.01021622,
      "balance_loss_clip": 0.93410671,
      "balance_loss_mlp": 1.01443696,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 1.6557812203111701,
      "language_loss": 0.71995556,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74176645,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.6490910053253174
    },
    {
      "auxiliary_loss_clip": 0.01161099,
      "auxiliary_loss_mlp": 0.01023798,
      "balance_loss_clip": 0.9725979,
      "balance_loss_mlp": 1.01672876,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 2.6126214338611264,
      "language_loss": 0.76195955,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.78380847,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 2.6183433532714844
    },
    {
      "auxiliary_loss_clip": 0.01168754,
      "auxiliary_loss_mlp": 0.01026571,
      "balance_loss_clip": 0.93105996,
      "balance_loss_mlp": 1.01917386,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 1.8827449172045831,
      "language_loss": 0.75090969,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77286291,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.6546480655670166
    },
    {
      "auxiliary_loss_clip": 0.01157667,
      "auxiliary_loss_mlp": 0.01022562,
      "balance_loss_clip": 0.89442396,
      "balance_loss_mlp": 1.0152787,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 1.9607965581614577,
      "language_loss": 0.71075457,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73255688,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.6624186038970947
    },
    {
      "auxiliary_loss_clip": 0.01058651,
      "auxiliary_loss_mlp": 0.01002385,
      "balance_loss_clip": 1.00797868,
      "balance_loss_mlp": 1.00075173,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7976533486050813,
      "language_loss": 0.5581919,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57880223,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.2308576107025146
    },
    {
      "auxiliary_loss_clip": 0.01166023,
      "auxiliary_loss_mlp": 0.0102784,
      "balance_loss_clip": 0.97149068,
      "balance_loss_mlp": 1.02041936,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 1.9891956612560886,
      "language_loss": 0.69088662,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71282518,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.671172857284546
    },
    {
      "auxiliary_loss_clip": 0.01166803,
      "auxiliary_loss_mlp": 0.01032893,
      "balance_loss_clip": 0.89355141,
      "balance_loss_mlp": 1.02617288,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.268638413673641,
      "language_loss": 0.70954317,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.73154014,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 3.5174410343170166
    },
    {
      "auxiliary_loss_clip": 0.0115803,
      "auxiliary_loss_mlp": 0.01025483,
      "balance_loss_clip": 1.0056107,
      "balance_loss_mlp": 1.01775837,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 1.862942456158555,
      "language_loss": 0.84098089,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86281598,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.5854506492614746
    },
    {
      "auxiliary_loss_clip": 0.01162479,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.01093185,
      "balance_loss_mlp": 1.01956296,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 1.9476591935110805,
      "language_loss": 0.71390468,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73579431,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 2.6518497467041016
    },
    {
      "auxiliary_loss_clip": 0.01164435,
      "auxiliary_loss_mlp": 0.01022921,
      "balance_loss_clip": 1.00955558,
      "balance_loss_mlp": 1.01549387,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 1.6776806857995483,
      "language_loss": 0.62930238,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.65117586,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.727675437927246
    },
    {
      "auxiliary_loss_clip": 0.01164253,
      "auxiliary_loss_mlp": 0.01021404,
      "balance_loss_clip": 1.01061201,
      "balance_loss_mlp": 1.01437688,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 2.5937989129828862,
      "language_loss": 0.87135011,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89320666,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.793147087097168
    },
    {
      "auxiliary_loss_clip": 0.01169364,
      "auxiliary_loss_mlp": 0.01028816,
      "balance_loss_clip": 0.96937704,
      "balance_loss_mlp": 1.02162504,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 7.223355164932529,
      "language_loss": 0.77658606,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.79856783,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 3.6463546752929688
    },
    {
      "auxiliary_loss_clip": 0.01169272,
      "auxiliary_loss_mlp": 0.01026352,
      "balance_loss_clip": 1.04861856,
      "balance_loss_mlp": 1.01918113,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 1.9783157692104834,
      "language_loss": 0.71830761,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74026382,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.5482873916625977
    },
    {
      "auxiliary_loss_clip": 0.01161089,
      "auxiliary_loss_mlp": 0.01023115,
      "balance_loss_clip": 0.93421394,
      "balance_loss_mlp": 1.01642489,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 2.156409276759633,
      "language_loss": 0.72941518,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.75125718,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.7635416984558105
    },
    {
      "auxiliary_loss_clip": 0.01160989,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 0.8906526,
      "balance_loss_mlp": 1.01734638,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 2.066142949639567,
      "language_loss": 0.79821813,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.82007426,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 3.5747299194335938
    },
    {
      "auxiliary_loss_clip": 0.01158698,
      "auxiliary_loss_mlp": 0.01034874,
      "balance_loss_clip": 0.93348217,
      "balance_loss_mlp": 1.02721453,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 2.0145356268601473,
      "language_loss": 0.79387248,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.81580818,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 3.5206480026245117
    },
    {
      "auxiliary_loss_clip": 0.01058594,
      "auxiliary_loss_mlp": 0.01002147,
      "balance_loss_clip": 1.00798726,
      "balance_loss_mlp": 1.00051427,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8309940912371861,
      "language_loss": 0.62838948,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64899695,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 3.050913095474243
    },
    {
      "auxiliary_loss_clip": 0.01166783,
      "auxiliary_loss_mlp": 0.01028107,
      "balance_loss_clip": 0.97539806,
      "balance_loss_mlp": 1.02066278,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 2.7452692551170443,
      "language_loss": 0.68630064,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.70824957,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.6107382774353027
    },
    {
      "auxiliary_loss_clip": 0.01166343,
      "auxiliary_loss_mlp": 0.01026242,
      "balance_loss_clip": 0.97106886,
      "balance_loss_mlp": 1.01926816,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 3.051929357036058,
      "language_loss": 0.88430822,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90623403,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 2.627230405807495
    },
    {
      "auxiliary_loss_clip": 0.0106568,
      "auxiliary_loss_mlp": 0.01003494,
      "balance_loss_clip": 0.93338722,
      "balance_loss_mlp": 1.00170636,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7693867517206213,
      "language_loss": 0.57808965,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59878135,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.187525749206543
    },
    {
      "auxiliary_loss_clip": 0.01169055,
      "auxiliary_loss_mlp": 0.01022752,
      "balance_loss_clip": 1.04888391,
      "balance_loss_mlp": 1.01533723,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 3.539751562039569,
      "language_loss": 0.72465587,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.74657393,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.5834240913391113
    },
    {
      "auxiliary_loss_clip": 0.01163303,
      "auxiliary_loss_mlp": 0.01025213,
      "balance_loss_clip": 0.97350234,
      "balance_loss_mlp": 1.01768851,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 1.4660574793606551,
      "language_loss": 0.66502494,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.68691015,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 2.713655948638916
    },
    {
      "auxiliary_loss_clip": 0.01163229,
      "auxiliary_loss_mlp": 0.01021106,
      "balance_loss_clip": 1.00864458,
      "balance_loss_mlp": 1.01398349,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 2.2003122855184363,
      "language_loss": 0.73798525,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.75982857,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 2.6131551265716553
    },
    {
      "auxiliary_loss_clip": 0.01154199,
      "auxiliary_loss_mlp": 0.01028109,
      "balance_loss_clip": 1.00640416,
      "balance_loss_mlp": 1.02112889,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 1.6649587389392542,
      "language_loss": 0.79165602,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81347907,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 2.659653425216675
    },
    {
      "auxiliary_loss_clip": 0.01149129,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 0.89307189,
      "balance_loss_mlp": 1.01986718,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 1.5850859284159626,
      "language_loss": 0.632581,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.65434259,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 2.8030622005462646
    },
    {
      "auxiliary_loss_clip": 0.01144733,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 0.92709476,
      "balance_loss_mlp": 1.01925516,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 2.264777289805609,
      "language_loss": 0.78013581,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80185109,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.72644305229187
    },
    {
      "auxiliary_loss_clip": 0.01155415,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 0.9327054,
      "balance_loss_mlp": 1.02125061,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 1.7475372961672306,
      "language_loss": 0.8360945,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.85793352,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.691899538040161
    },
    {
      "auxiliary_loss_clip": 0.01173167,
      "auxiliary_loss_mlp": 0.01029492,
      "balance_loss_clip": 1.01287079,
      "balance_loss_mlp": 1.02192187,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 2.251128376837346,
      "language_loss": 0.7791574,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.80118406,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.661508798599243
    },
    {
      "auxiliary_loss_clip": 0.01162052,
      "auxiliary_loss_mlp": 0.01029243,
      "balance_loss_clip": 0.96890157,
      "balance_loss_mlp": 1.0221324,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 3.2051449451667553,
      "language_loss": 0.73267746,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.75459039,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 2.665259599685669
    },
    {
      "auxiliary_loss_clip": 0.01164183,
      "auxiliary_loss_mlp": 0.01023687,
      "balance_loss_clip": 0.97188544,
      "balance_loss_mlp": 1.01686227,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 1.7337527676020181,
      "language_loss": 0.73823845,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76011717,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.691427230834961
    },
    {
      "auxiliary_loss_clip": 0.01166958,
      "auxiliary_loss_mlp": 0.01024188,
      "balance_loss_clip": 0.97242594,
      "balance_loss_mlp": 1.01704121,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 1.968924370125575,
      "language_loss": 0.79148412,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.81339556,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.72015643119812
    },
    {
      "auxiliary_loss_clip": 0.0116649,
      "auxiliary_loss_mlp": 0.01027871,
      "balance_loss_clip": 1.04672956,
      "balance_loss_mlp": 1.02067316,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 1.4721463305882783,
      "language_loss": 0.74422157,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.76616514,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.6334171295166016
    },
    {
      "auxiliary_loss_clip": 0.01165207,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 0.89153701,
      "balance_loss_mlp": 1.01980972,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 2.214245808027268,
      "language_loss": 0.75829619,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.78021467,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 3.638455867767334
    },
    {
      "auxiliary_loss_clip": 0.01162533,
      "auxiliary_loss_mlp": 0.01030692,
      "balance_loss_clip": 0.97097754,
      "balance_loss_mlp": 1.02375412,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 2.024375291831616,
      "language_loss": 0.62008607,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.64201832,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.8497397899627686
    },
    {
      "auxiliary_loss_clip": 0.01165296,
      "auxiliary_loss_mlp": 0.01021938,
      "balance_loss_clip": 1.04668891,
      "balance_loss_mlp": 1.01542926,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 3.0324716757814993,
      "language_loss": 0.6996752,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72154748,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 2.584022283554077
    },
    {
      "auxiliary_loss_clip": 0.01165848,
      "auxiliary_loss_mlp": 0.01024764,
      "balance_loss_clip": 1.0080986,
      "balance_loss_mlp": 1.01797175,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.2966204659285996,
      "language_loss": 0.81380367,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83570981,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 2.5421054363250732
    },
    {
      "auxiliary_loss_clip": 0.01170085,
      "auxiliary_loss_mlp": 0.01027731,
      "balance_loss_clip": 0.93377125,
      "balance_loss_mlp": 1.02048278,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 1.865592856267681,
      "language_loss": 0.6368469,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.65882504,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.635129451751709
    },
    {
      "auxiliary_loss_clip": 0.01164071,
      "auxiliary_loss_mlp": 0.01026445,
      "balance_loss_clip": 0.93488008,
      "balance_loss_mlp": 1.01883674,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 1.523760286089498,
      "language_loss": 0.76875007,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79065526,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 3.923131227493286
    },
    {
      "auxiliary_loss_clip": 0.01117112,
      "auxiliary_loss_mlp": 0.01012033,
      "balance_loss_clip": 0.99012351,
      "balance_loss_mlp": 1.00749719,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.2048408670407436,
      "language_loss": 0.55020851,
      "learning_rate": 0.0,
      "loss": 0.57149994,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.1828629970550537
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7894440462679287,
      "train_runtime": 25495.1894,
      "train_samples_per_second": 13.047,
      "train_steps_per_second": 0.326
    }
  ],
  "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
}